University of Vermont ScholarWorks @ UVM
UVM College of Arts and Sciences College Honors Theses Undergraduate Theses
2015
Explaining Support for Authoritarianism in New Democracies
Andrew G. Goss University of Vermont
Follow this and additional works at: https://scholarworks.uvm.edu/castheses
Recommended Citation Goss, Andrew G., "Explaining Support for Authoritarianism in New Democracies" (2015). UVM College of Arts and Sciences College Honors Theses. 18. https://scholarworks.uvm.edu/castheses/18
This Undergraduate Thesis is brought to you for free and open access by the Undergraduate Theses at ScholarWorks @ UVM. It has been accepted for inclusion in UVM College of Arts and Sciences College Honors Theses by an authorized administrator of ScholarWorks @ UVM. For more information, please contact [email protected].
Explaining Support for Authoritarianism in New Democracies
Andrew Goss April 17, 2015
1 “Democracy is the worst form of government, except for all those other forms that have been tried from time to time.” -Winston Churchill
Chapter 1: Introduction
In the last few decades, significant democratic advances have been made all across the world. A body of research on democratization has emerged to explain this phenomenon.
However, one important question is often overlooked in the rush to try to explain the global democratic push: do the citizens of these democratizing countries actually want democracy?
Often the answer seems to be yes. Perhaps because of universally accepted norms, positive connotations with the word itself, or dreams of economic success, democracy retains a large margin of support compared to authoritarian alternatives across the world, especially when investigated in simple terms. However, this support is not uniform and falters in some new democracies more than others. Surveys in several of these countries have revealed that surprisingly large constituencies still view an authoritarian government as preferable.
This thesis will investigate the puzzle demonstrated in Chart 1.1, which shows the percentage of respondents across Latin America and Africa who agreed that authoritarianism, or a “non-democratic regime,” can be preferable to democracy. No country exceeded 30% support for authoritarianism—good news for democracy—but there is significant variance across different countries. Average authoritarian support in Latin America is 18.4% and ranges from
11.9% in Bolivia to 28.9% in Honduras. In Africa, authoritarian support is generally lower. The
African average is 11%, less than even the lowest Latin American country. However, there is still notable variance across the region and Lesotho presents itself as a significant outlier with
29.7% of survey takers agreeing that an authoritarian regime could be preferable. The puzzle
2 that emerges in Chart 1.1 is: what explains the wide variation between countries in levels of popular support for authoritarianism?
Chart 1.1: Authoritarian Support in Africa and Latin America
Sources: Afrobarometer 2008/2009, Latinobarómetro 2011
The puzzle is also apparent when we examine patterns of authoritarian support in Central and Eastern Europe (CEE). To investigate this region , the best available source is the World
Values Survey (WVS). We can gain a relevant glimpse into authorit arian support based on the percentage of survey takers who reject (view as fairly bad or very bad) both military rule and rule by a strong leader who “does not have to deal with parliament or elections.” In Romania and
Ukraine, the level of rejection is extremel y low: 18.8% and 27% respectively. In Poland and
Estonia, the rate of rejection is hi gher: 56% and 58.6%. While the number of country cases is limited, it is convenient for analysis that the four countries divide neatly into categories of high and low authoritarian support.
3 Using similar measures of support for concrete regime alternatives in Latin America and
Africa shows even greater variation in support for authoritarian regime types in these regions.
29.2% of respondents across 18 countries in Latin America agreed that they “would support a
military government in replacement of a democratic one if things got bad.” Conditional support
for military rule reached as high as 56.6% in Guatemala and as low as 9% in Costa Rica. In
Africa, support for one-party rule reached 39.2%, 36.2%, and 34.1% in Lesotho, Mozambique
and Burkina Faso respectively. The same measure found only 4.9% support in Senegal and less than a tenth in Madagascar. Appendix A lists these various measures of authoritarian support across the three regions observed in this study.
These trends highlight the significant variation among different new democracies with regards to their support for the idea of an authoritarian regime. These measures of authoritarian support will serve as the dependent variable in the following investigation of the factors that might contribute to the differences observed. I will examine why authoritarian support remains a key feature in the political makeup of many of these countries despite the adoption of at least nominally democratic institutions. Beginning with the relevant literature, I will investigate factors that might be causing the persistence or emergence of support for non-democratic regime alternatives.
Literature Review and Hypotheses
The major thrust of academic work related to this particular issue has addressed the
causes of support for democracy rather than support for authoritarianism . In many ways support
for democracy (or democratic legitimacy) is the opposite side of the same coin and can offer
clues for this project. However, there are important differences between these two measures of
support. For example, a lack of democratic support could indicate widespread indifference to
4 regime type rather than an embrace of some concrete non-democratic alternative. Measures of support for authoritarianism also tend to show larger disparities between countries and can therefore guide a more careful and nuanced investigation. In this project, I am interested in investigating the root sources of popular support for the aim of reintroducing authoritarian institutions into newly democratic countries. Since the division between these two research questions is not well defined and often overlapping, the following review of the literature will draw on both measures of regime support in order to form hypotheses.
Explanations for regime support tend to focus on economic, political and cultural factors.
One classic line of scholarship has emphasized the role of economic factors in determining support for democracy. Writing in the 1950s Seymour Martin Lipset made the case that modernization would promote democratization. 1 His theory of modernization suggests that through industrialization, urbanization, and widening of education, democratic values would develop among the people. Later scholars have stressed the importance of the more immediate economic experience of new democracies in shaping attitudes toward democracy. 2 Proponents of this argument expect to find support for democracy in states that achieve growth and wealth under democratic institutions. Conversely, an experience of hardship and scarcity should decrease commitment to democracy. This school of thought helps generate the following hypothesis regarding support for authoritarianism: poor and/or stagnant economic performance under democratic institutions will cause disenchantment with democracy and a consequent preference for authoritarian rule .
1 Seymour Martin Lipset, “Some Social Requisites of Democracy: Economic Development and Political Legitimacy,” The American Political Science Review (1959), 69-105. 2 Adam Przeworski, Democracy and the Market: Political and Economic Reforms in Eastern Europe and Latin America (New York: Cambridge University Press, 1991); Herbert Kitschelt, “The Formation of Party Systems in East Central Europe,” Politics and Society (1992), 7-50.
5 The investigation of these factors, carried out in Chapter 2, questions this hypothesis, drawing from and expanding on the literature presented above. I measure levels of education, urbanization, GDP per capita, GDP growth, inflation, economic inequality and poverty in countries with high popular authoritarian support as compared to those with lower support.
Additionally, seeking to fully grasp the economic realities for citizens in these countries, I look at the effect of survey takers’ assessments of their countries’ and their own household economic situations on regime support.
Many theorists warn against putting too much stock into a country’s economic experience to explain support for democracy and authoritarian alternatives. Numerous studies looking at a wide variety of possible explanations suggest that political variables rather than economic variables are most important.3 This body of research expects to find support for democracy and rejection of authoritarianism in states where democratic institutions have achieved rule of law, individual freedoms and rights, meaningful elections, and limits on corruption. Where this has failed, citizens will feel less committed to democracy and be more likely to embrace authoritarian options. The hypothesis that emerges from this literature is as follows: Failure to guarantee certain basic freedoms and securities associated with democratic institutions will drive citizens in new democracies to support some form of authoritarianism.
Chapter 3 investigates this hypothesis and the importance of the political system for determining authoritarian support. Again comparing countries with high authoritarian support to those with lower authoritarian support, this chapter looks at general measurements of politics in
3 Geoffrey Evans, Stephen Whitefield, “The Politics and Economics of Democratic Commitment: Support for Democracy in Transition Societies,” British Journal of Political Science (1995), 485-514; Richard Rose, William Mishler, Christian Haerpfer, Democracy and its Alternatives: Understanding Post-Communist Societies (Baltimore: John Hopkins University Press, 1998); Larry Diamond, Developing Democracy: Toward Consolidation , (Baltimore: The John Hopkins University Press, 1999).
6 each country as well as perceptions of corruption. Then, attempting to measure the experience of
individuals in these countries, I look for correlations between measures of authoritarian support
and several survey questions asking about democratic conditions on the ground. These include
worsening/improvement of democracy, the delivery of various political “goods” associated with
democracy, efficacy of the system, satisfaction with democracy, and corruption.
Chapter 3 also investigates the effect of current and past political violence stemming
from the regime in power. The hypothesis here is that an increase in repressive tactics used by
the government under democratic institutions (particularly with respect to past levels of
government violence) will lead some to be more open to a form of authoritarian rule similar to the seemingly safe past regime.
Culture is another factor evoked to explain differences in regime commitment. Some scholars have suggested that cultural factors like “Asian values” or religious traditions like Islam can produce a population that is more subservient to authority and willing to live under an illiberal regime. At least one study, looking at South Korea, finds that certain aspects of
Confucian cultural tradition drive support for authoritarian alternatives.4 Conversely, using
statistical analysis, Stephen Fish finds that devoutness of Muslims does not have a significant
effect on support for democracy.5 Proponents of the cultural argument would expect countries to
be bound by their dominant culture. In general this means that support for democracy would be
found in more liberal, Western countries while support for authoritarianism would be found in
countries with more hierarchical or traditional cultures. We can simplify this discussion with the
4 Chong-Min Park and Doh Chull Shin, “Do Asian Values Deter Popular Support For Democracy? The Case of South Korea,” Asian Barometer Working Paper Series, No. 26 (2004). 5 Steven M. Fish, Are Muslims Distinctive? A Look at the Evidence (New York: Oxford University Press, 2011).
7 following hypothesis: countries that are not characterized by a liberal, “Western” culture will generally be more likely to embrace authoritarian alternatives.
Using this hypothesis as a springboard, Chapter 4 attempts to quantify cultural differences between countries with varying levels of authoritarian support. I use Ronald
Inglehart and Christian Welzel’s index of emancipative values to evaluate the degree to which
Western, pro-democratic culture has penetrated a given society. Additionally, I investigate the proportion of survey-takers who see goals of security and economic development as more of a priority for their country than deepening democracy and building responsive institutions. Finally,
I compare levels of interpersonal trust across the new democracies observed.
Overview and Methods
The first section of this project will investigate the hypotheses enumerated above at the regional level for Africa, Latin America and Central and Eastern Europe. I will begin by examining the role of economics, politics, culture and history in the formation of preferences for authoritarian rule in these regions. The idea of focusing on the regional level has faced some recent resistance by those who believe that analyzing isolated regions could “fail to capture a significant cross-regional pattern.” 6 One could counter that a larger-scale analysis could similarly fail to identify a significant regional pattern. As a secondary inquiry, this project will seek to identify region-specific trends that may indicate that authoritarian support is better analyzed at this level.
Country cases were selected (and labeled democracies) from among those included in the relevant surveys if they received a score of at least partly free on Freedom House’s measure of
Freedom in the World during the year of the survey used for analysis. They will be divided into
6 Yun-han Chu, Michael Bratton, Marta Lagos, Sandeep Shastri, Mark Tessler, “Public Opinion and Democratic Legitimacy,” Journal of Democracy (2008), 76.
8 two groups based on levels of authoritarian support. In Latin America, the first group contains countries with over 20% saying that sometimes an authoritarian regime is preferable. The second group is made up of the remaining states where less than 20% of those surveyed accepted authoritarianism under some circumstances. 7 Due to the abundance of democracy-related
Afrobarometer questions and the comparatively small variation between cases, African countries were divided by a different measure of authoritarian support. Countries categorized as “more willing to accept authoritarianism” displayed less than 60% rejection of both strong-man and military rule as well as over 10% agreeing that sometimes an authoritarian regime can be better.
All other African countries were considered “less willing to accept authoritarianism.” 8 For these two regions, I will seek to find significant differences between both groups of countries in terms of economic, political and cultural/historical variables. In Central and Eastern Europe, the four countries will simply be presented individually.
During this regional analysis, I will also zoom in and scrutinize a specific country from each region with unusually high support for authoritarianism. These countries will be Guatemala,
Lesotho and Ukraine. These cases were chosen because of both their persistent support for authoritarianism over several survey periods as well as their high support based on several different measures (See Appendix A).
The second section of this project will investigate the peculiar case of high authoritarian support in Ukraine. I will then use the “most similar” systems (or method of difference)
7 Countries with high authoritarian support in Latin America: Brazil, Dominican Republic, Ecuador, Guatemala, Honduras, Panama, and Paraguay. Countries with low authoritarian support: Argentina, Bolivia, Chile, Colombia, Costa Rica, El Salvador, Mexico, Nicaragua, Peru, Uruguay, and Venezuela. 8 Countries with high authoritarian support in Africa: Burkina Faso, Lesotho, Malawi, Mali, Mozambique, Namibia, Nigeria, South Africa, and Tanzania. Countries with low authoritarian support: Benin, Botswana, Cape Verde, Ghana, Kenya, Liberia, Madagascar, Senegal, Uganda, and Zambia.
9 approach to root out the drivers of authoritarian sympathy in Guatemala as compared to
Nicaragua. This process involves choosing countries that are similar in many aspects but have different levels of support for authoritarianism. Due to the likeness of the cases in this comparative approach, I should be able to identify important differences between the two countries that may influence regime support. The findings from the regional analysis will guide these smaller-scale investigations.
Due to the fact that this investigation relies mainly on cross-national survey data, authoritarianism is conceptualized very generally as any non-democratic form of rule. Where survey questions simply ask respondents about authoritarianism as a regime choice, we should not expect all survey-takers to be applying the same definition. Often individual interpretations of what authoritarianism means will be determined by prior experience within the country or the experience of neighboring countries. Where possible, I will employ measures of support for concrete authoritarian alternatives where power is clearly not derived from the people, but rather from the military or a strongman ruler.
Significance
Many scholars have recognized the importance of legitimacy for the survival and development of any regime. Congruence theory is represented in a body of work arguing that political institutions in a country must be in line with similar orientations among the population. 9
That means that a democratic regime in a country with undemocratic or authoritarian citizens will be bound for failure. Larry Diamond makes this argument fairly convincingly in his book
Developing Democracy: Towards Consolidation, where he asserts that for a democracy to
9 Christian Welzel and Ronald F. Inglehart, “Political Culture, Mass Beliefs, and Value Change,” Democratization, edited by Christian W. Haerpfer, Patrick Bernhagen, Ronald F. Inglehart, and Christian Welzel (New York: Oxford University Press, 2009), 126-144.
10 consolidate, 70-75% of the population should consider democracy to be better than all other possible regime types. If this proves to be true, it would seem logical to look at the factors that drive this type of support. It would also seem particularly important to find countries that show wide support for those other possible regime types and understand why this is the case.
However, there may be some reason to doubt the idea that popular preferences will be borne out in political institutions. For example, the trends observed in Chart 1.1 above show higher levels of support for authoritarianism in Latin America as compared to Africa even though levels of democracy are generally higher in the former region. Even if one does not buy the argument that authoritarian-minded citizens could be a danger for the development of democracy, this investigation still has value. The question remains disputed within the democratization literature. The persistence of support for authoritarianism is an interesting puzzle that can help improve an understanding of the way that countries transition to democracy and the role of the people in that process. This research project will attempt to add to what I believe to be an understudied but important factor in movements toward democracy.
11 Appendix A: Latinobarómetro (2011) Freedom % Agreeing that % Disagreeing % Agreeing “I Average House “under some that “Democracy would support a Score circumstances, may have military government (2010) an authoritarian problems but it is in replacement of a government can still the best form democratic one if be preferable” of government” things got very bad” Argentina 4 (F) 13.8 10.5 24.4 16.2 Bolivia 6 (PF) 11.9 24.2 21.2 19.1 Brazil 4 (F) 22.3 17.5 24.4 21.4 Chile 2 (F) 14.6 11.7 20.0 15.4 Colombia 7 (PF) 12.1 17.6 32.7 20.8 Costa Rica 2 (F) 15.1 17.2 9.0 13.8 Dominican 4 (F) 22.5 16.4 23.8 20.9 Republic Ecuador 6 (PF) 24.4 14.2 24.6 21.1 El Salvador 5 (F) 17.4 24.2 35.6 25.7 Guatemala 8 (PF) 24.7 36.7 56.6 39.3 Honduras 8 (PF) 28.9 24.1 37.3 30.1 Mexico 6 (PF) 15.4 40.9 41.0 32.4 Nicaragua 8 (PF) 17.8 16.2 19.9 18.0 Panama 3 (F) 21.4 14.6 26.5 20.8 Paraguay 6 (PF) 26.2 19.1 43.8 29.7 Peru 5 (F) 17.4 21.7 41.6 26.9 Uruguay 2 (F) 12.0 6.3 19.2 12.5 Venezuela 10 (PF) 14.1 12.7 25.9 17.6
World Values Survey (2012-2014) % Saying Good: % Saying Good: % Saying Average A Strong Leader Having the army democracy is not who does not rule very important have to bother (1-4 on a 1-10 with parliament scale) or Elections Chile 37.4 17.9 4 19.8 Colombia 55.9 33.8 7.2 32.3 Ecuador 71.3 24.7 1.8 32.6 Mexico 58.7 52.6 2.9 38.1 Peru 60.4 28.1 6.5 31.7 Uruguay 39.3 9.2 4.6 17.7
12
Afrobarometer (2008/2009) Freedom % % % % Agreeing Average House Approving Approving Approving that Score of one-party of military of one-man sometimes a (2008) rule rule rule non- democratic government can be preferable Benin 4 (F) 15.4 17.2 6.4 9.8 12.2 Botswana 4 (F) 17.7 8.3 4.4 4.8 8.8 Burkina Faso 8 (PF) 34.1 38.9 15 10.8 24.7 Cape Verde 2 (F) 11.2 6.2 11.4 6.3 8.8 Ghana 3 (F) 15 16 9.5 7.1 11.9 Kenya 6 (PF) 14.2 3 4 7.8 7.3 Lesotho 5 (F) 39.2 17.2 17.1 29.7 25.8 Liberia 8 (PF) 16.3 19.4 8.9 9.1 13.4 Madagascar 6 (PF) 9.7 12.6 7.1 5.7 8.8 Malawi 8 (PF) 29.3 9.8 15.5 12.4 16.8 Mali 4 (F) 15.8 24.4 10.4 11.7 15.6 Mozambique 7 (PF) 36.2 18.4 23 15.5 23.3 Namibia 4 (F) 22.6 18.3 15.6 10.7 16.8 Nigeria 8 (PF) 15.9 12.4 8.3 16.8 13.4 Senegal 5 (F) 4.9 19.4 4.1 6.4 8.7 South Africa 3 (F) 19.5 14.7 13.6 17.8 16.4 Tanzania 7 (PF) 34.4 7.9 5.7 12.7 15.2 Uganda 9 (PF) 16.1 12.5 5.7 6.7 10.3 Zambia 8 (PF) 16.5 6.3 4.7 7 8.6 Zimbabwe 13 (NF) 28.1 14.7 8.9 6.1 14.5
13
Afrobarometer (2010/2011) Freedom % % % % Agreeing Average House Approving Approving Approving that Score of one-party of military of one-man sometimes a (2011) rule rule rule non- democratic government can be preferable Benin 4 (F) 10 11 5 9 8.8 Botswana 5 (F) 19 8 6 7 10 Burkina Faso 7 (PF) 15 24 6 5 12.5 Burundi 10 (PF) 19 12 11 3 11.3 Cape Verde 2 (F) 9 9 5 6 7.3 Ghana 3 (F) 9 11 6 7 8.3 Guinea 10 (PF) 13 23 20 10 16.5 Kenya 7 (PF) 22 9 6 10 11.8 Lesotho 6 (PF) 32 14 11 19 19 Liberia 7 (PF) 7 17 6 5 8.8 Madagascar 10 (PF) 10 20 8 9 11.8 Malawi 7 (PF) 23 10 11 14 14.5 Mali 5 (F) 18 34 13 19 21 Mauritius 3 (F) 5 2 3 5 3.75 Morocco 9 (PF) 6 6 6 10 7 Mozambique 7 (PF) 25 17 13 11 16.5 Namibia 4 (F) 20 15 13 15 15.75 Niger 7 (PF) 15 30 19 8 18 Nigeria 8 (PF) 10 14 7 19 12.5 Senegal 6 (PF) 9 13 8 5 8.75 Sierra Leone 6 (PF) 10 9 6 15 10 South Africa 4 (F) 21 15 15 15 16.5 Togo 9 (PF) 12 28 10 5 13.8 Tunisia 7 (PF) 32 33 12 11 22 Uganda 9 (PF) 15 8 6 10 9.8 Tanzania 6 (PF) 22 18 6 10 14 Zambia 7 (PF) 12 3 4 4 5.8
World Values Survey 2010-2014 % Approving of % Approving of % Disapproving Average strongman rule military rule of democracy Estonia 29.2 3.9 12.3 15.1 Poland 20 19 14.9 18.0 Romania 69.7 28.6 11.9 36.7 Ukraine 71.3 12.7 14.6 32.9
14 Chapter 2: Economics and Regime Support
An early line of scholarship, generally looking at the transitions to democracy in Central and Eastern Europe (CEE), stresses the importance of a new democracy’s economic experience in shaping its population’s regime preferences. Some proponents of this view highlight the importance of modernization-related factors like education and urbanization. Others tend to view regime preference as a simple economic consideration: would I (or my country) be better off economically under a different regime type? The expectation is to find support for democracy in states that achieve a net improvement in the economic situation of the country. In general, an experience of growth and wealth under democratic institutions should promote democratic support while an experience of hardship or scarcity should cause a desire to turn to some authoritarian alternative.
Looking at the transitions in CEE, Adam Przeworski suggests that the populations of transitioning countries have little experience with a market economy and form expectations based on promises from political leaders. Their assessment of democracy as a regime then will be determined by how the reality of democracy and a market economy stack up to their expectations. 10 Herbert Kitschelt also offers a perspective on this issue in an article explaining the formation of party systems in CEE. He argues that countries that experience less economic development will have populations that are more authoritarian in their political orientations and preferred parties. Conversely, more economic development will produce a more libertarian, inclusive and participatory population. 11
10 Adam Przeworksi, Democracy and the Market (Cambridge: Cambridge University Press, 1991), 184. 11 Herbert Kitschelt. “The Formation of Part Systems in East Central Europe”, Politics and Society , 20 (1992), 19-20.
15 Proponents of this line of thinking give a number of theoretical reasons for the connection
between regime support and economics. These tend to take the form of modernization
explanations and explanations rooted in more short-term economic experiences. Those viewing
modernization as central to regime support claim that increasing affluence, a larger middle class,
more free time, and better educational opportunities will “democratize” the population. As
Kitschelt puts it, these conditions will increase citizens’ capacities to “exhibit individualism and
to claim greater control over their life sphere.” 12 Others have argued that development and
wealth provides citizens with a stake in the system, making politics less of a zero-sum game. 13
Those who find the recent economic experience of a country to be more important tend to
fall closer to Przeworski’s framework that regime support will be determined by how the reality
of democracy stacks up to citizens’ expectations. Following this argument, citizens will judge
democracy based on whether the government is able to provide an improvement in living
standards or if, instead, economic conditions cause uncertainty and suffering. 14 Additionally,
Economic performance in the short-term has been found to drive satisfaction with democracy,
which is in turn correlated with support for democracy, suggesting at least an indirect effect on
regime legitimacy. 15
Applying these ideas to authoritarian support, we may first expect to see high popular
willingness to accept authoritarian institutions of government in less modernized countries
characterized by lower per capita GDP, education and levels of urbanization. Perhaps in these
12 Ibid, 20. 13 Geoffrey Evans and Stephen Whitefield, “The Politics and Economics of Democratic Commitment: Support for Democracy in Transition Societies”, British Journal of Political Science, Vol. 25, No. 4 (1995), 490. 14 Ibid. 15 Larry Diamond, Developing Democracy: Toward Consolidation (Baltimore: John Hopkins University Press, 1999), 203.
16 less developed countries, authoritarian values persist because they have not had the
democratizing experience of industrialization, urbanization and education. Turning to short-term
explanations, we might expect that in a situation of deprivation, high inflation, and poverty,
citizens might blame the political system and shift their loyalty to some non-democratic
alternative. 16 This assumes that citizens will view some kind of authoritarian regime as more stable, decisive and better armed to tackle difficult economic issues. Indeed the process of economic reform under democratic institutions sacrifices some degree of efficiency for a more inclusive, open decision-making process.
In order to get a full picture of the differences in economic situations between countries, I look at both objective and subjective measures of how the economy is performing. First, I will investigate a series of objective (or macro-level) measures of both modernization and economic growth such as GDP per capita, GDP growth, urbanization, education, inflation, inequality and poverty, in order to capture each country’s recent and historical economic experience under democracy. In the second half of this chapter I will look at several subjective (or micro-level) measures of perceptions of economic conditions derived from responses to survey questions contained in the Latinobarómetro, Afrobarometer and World Values Survey.
Objective Measures
The tables below include several measures that seek to capture important elements of the economic situation in the regions under investigation. To measure levels of modernization and development, I have employed levels of GDP per capita, percentage of the population that is urban, and mean years of education. All of these variables are expected to be lower in countries characterized by higher popular support for authoritarianism. To capture each country’s more
16 Ibid.
17 recent economic experience under democratic institutions, I rely on ten-year averages of GDP
growth and inflation. Lower GDP growth and higher inflation are expected to be associated with
countries that display higher levels of authoritarian support, as citizens will be disappointed in
the democratic government’s inability to provide prosperity and may seek alternative means to
do so. Related to this is a measure of the percent of the population below the national poverty
line in each country. This seeks to capture the degree to which democracy appears able to tackle
the difficult issue of its impoverished citizens. Higher poverty should be associated with a
willingness to accept non-democratic solutions in the form of a more authoritarian regime.
Poverty might also be linked to modernization and education levels, which could have an
independent effect on regime support.
The final measure employed below looks at levels of inequality captured by each state’s
GINI coefficient. 17 Although the scholarship does not generally indicate inequality as a driver of authoritarian or democratic support, there may be some reason to believe it is relevant. On the one hand, citizens might blame the current democratic government for failing to address this important economic issue, shifting their hopes to an authoritarian populist promising speedy reform. Alternatively, deepening democracy could threaten elite interests and turn some who are on the winning side of inequality toward support for authoritarian options. My analysis indicates a small but notable effect of inequality on authoritarian support in Latin America, but the drivers behind this trend are somewhat unclear.
To determine whether these factors contribute to authoritarian support, the countries have been divided into two categories based on the criteria identified in Chapter 1. The following analysis will look for significant differences between the group of countries with populations
17 The GINI index measures how far the distribution of income in a particular country is from perfect equality. On a 0-100 scale, a higher score indicates a more unequal country.
18 identified as more willing to support authoritarianism and those that are less willing.
Additionally, I will separate out one interesting case in both Latin America and Africa that has notably high levels (Guatemala and Lesotho).
The results suggest that there is significant regional variation with regard to the salience of economic factors. In Latin America, modernization and inequality appear somewhat relevant to explanations of authoritarian support. In CEE, poor short-term economic experience and low
GDP per-capita characterize the countries with higher authoritarian support. In Africa, there is little evidence that economic factors are important at all.
Latin America
In Latin America, Table 2.1 shows some interesting differences between countries with higher authoritarian support and those with lower levels. In terms of modernization, the differences are as predicted with higher authoritarian support being present in countries with a smaller urban population, lower GDP per capita and lower levels of education. However, only urbanization is significant at the 0.1 level. In general, this offers some tentative support to the idea that countries that are slow to develop and industrialize may be prone to higher authoritarian support. An experience of urbanization could introduce more open, democratic ideas to a population as opposed to a more isolated and traditional rural environment. Still, a lack of development is not the death knell of popular authoritarian sentiment. Some countries, like
Brazil and Panama, where authoritarian options are more popular, have levels of modernization comparable to there less authoritarian counterparts. Additionally, Bolivia and Nicaragua are characterized by relatively low authoritarian support despite some of the lowest levels of per capita GDP and urbanization.
19 Table 2.1 generally indicates little difference between the two sets of countries in terms of inflation, GDP growth and poverty level. This contradicts the predictions offered by the literature reviewed above. In this region, it appears that authoritarian support can thrive under both good and bad economic conditions. Based on this data, poor short-term economic performance does not seem to be fueling democratic defectors.
Table 2.1: Economics and Authoritarian Support in Latin America Guatemala Countries With Countries With Difference (With High Higher Lower (two-tailed p- Authoritarian Authoritarian Authoritarian value) Support) Support Support (St. Dev.) N=7 (St. Dev.) N=11 Modernization Urban Population 50% 64.14% 77.9% -13.76% (% of Total) (12.44) (11.94) (0.06) GDP Per Capita 3478 6309 9522 -3213 (USD) (3535) (5327) (0.18) Mean Years of 5.3 7.16 8.24 -1.08 Schooling (1.40) (1.32) (0.16) (Adults) Recent Economic Experience Average GDP 3.44 4.43 3.73 0.70 Growth 2001- (1.14) (1.09) (0.24) 2011 Average Inflation* 6.77 7.50 6.01 1.49 2001-2011 (2.83) (2.64) (0.33) GINI Coefficient 52.4 51.56 46.61 4.95 ~2011 (3.75) (3.49) (0.03) Proportion Living .54 .366 .323 .043 Under National (0.15) (0.11) (0.54) Poverty Line *3 countries are missing data for this variable Sources: The World Bank, UN Human Development Reports, Latinobarómetro 2011
Moving beyond these basic economic indicators, there is an interesting difference between the GINI coefficients of both groups of countries indicating that, in Latin America, more unequal countries have populations that are more supportive of authoritarianism. This relationship is demonstrated in Chart 2.1 below. The two groups of countries have, on average,
20 about a 5-point difference on the GINI index. One possible interpretation of this is that the centrality of land reform in many emerging Latin American democracies combined with the difficulty of carrying it out under those new democratic institutions may cause populations to seek out alterna tive regime possibilities. 18 Alternatively, this gap could be caused by a fear among the entrenched elite that democracy might bring a more equal distribution of wealth and power. While I do not find decisive evidence in either direction, results from sur vey data presented further on seem to suggest that in some cases it is more likely the latter explanation that fits in Latin America.
Chart 2. 1: Inequality vs Willingness to Accept Authoritarianism in Latin America
Sources: World Bank, Latinobarómetro 2 011
18 Diamond, Developing Democracy, 84.
21 Guatemala sits at the negative extreme of each measure of economic experience. It is characterized by an undeveloped economy, extremely high poverty and inequality, as well as relatively low GDP growth. Whether these extreme levels have a greater effect on regime support than the slighter differences observed across other countries will be explored with a closer look at survey data further on. In general I find that this is not the case.
Africa
In Africa, Table 2.2 shows no notable relationship between either modernization or economic growth indicators and authoritarian support. For all variables the difference between both groups was miniscule. In the cases of GDP per capita, GDP growth, inflation and poverty the data actually shows poorer conditions among the group of countries less supportive of authoritarianism.
Lesotho is somewhat unique in that it has lower GDP growth and greater inequality than the regional average. Again, the degree to which this could indicate that relatively extreme economic stagnation has a greater effect on regime preference than smaller differences between other countries will be observed using subjective measures. However, the data as a whole seems to contradict the idea that economic factors play a role in forming opinions about authoritarian regime alternatives in Africa. This is mostly consistent with the findings of similar studies on the region. 19
19 Michael Bratton and Robert Mattes, “Support for Democracy in Africa: Intrinsic or Instrumental?” British Journal of Political Science 31(3) (2001) 447-474.
22 Table 2.2: Economics and Authoritarian Support in Africa Variable Lesotho Countries With Countries With Difference (With High Populations more Populations less (p-value) Support for Willing to Willing to Authoritarianism) Accept Accept Authoritarianism Authoritarianism (St. Dev.) N=9 (St. Dev.) N=11 Modernization Urban 25% 35% 41.3% -6.3% Population (% of (13.74) (14.49) (0.36) Total) GDP Per Capita 1126 2152 1937 214.81 (USD) (2416) (2136) (0.84) Mean Years of 5.9 4.76 5.43 -0.67 Schooling (2.51) (1.75) (0.52) (Adults) Recent Economic Experience Average GDP 3.8 5.60 4.48 1.12 Growth 2000- (1.64) (4.30) (0.45) 2008 Average 7.79 7.82 9.18 -1.36 Inflation 2000- (4.22) (5.85) (0.58) 2008* GINI coefficient 51.5 46.48 44.67 1.81 ~2006*# (12.05) (7.51) (0.71) Proportion .49 .458 .461 -.013 Living Under (0.13) (0.18) (0.86) National Poverty Line *excluding zimbabwe #data used was the closest to 2006 available Sources: Afrobarometer 2008/2009, The World Bank
Central and Eastern Europe
While it is difficult to make any definite conclusions based on the small number of cases available for Central and Eastern Europe, the table below reveals some interesting trends. The more authoritarian-leaning cases (Ukraine and Romania) demonstrate notably worse economic performance than their less-authoritarian counterparts (Estonia and Poland). Ukraine and
Romania have lower average GDP growth over the ten years prior to the survey, a higher ten-
23 year average of inflation and a higher proportion of the population living under the national poverty line. These findings are in line with the expectations of previous scholars looking at the region and suggest that an experience of economic hardship might lead some to reject democracy for some illiberal alternative. Still, a 0.3 difference (at the highest) between GDP growths in these countries does not suggest a huge effect of economics on regime preference. Oddly,
Inequality was actually higher in the cases with less support for authoritarianism.
Table 2.3: Economics and Authoritarian Support in Central and Eastern Europe More Willing to Accept Less Willing to Accept Authoritarian Alternatives Authoritarian Alternatives Country Ukraine Romania Poland Estonia (% Rejection of (27% rejection) (18.8% rejection) (56% rejection) (58.6% rejection) both one-man and military rule) Modernization Urban Population 69% 54% 61% 68% (% of Total) GDP Per Capita 3901 9499 13,648 18,783 (USD) Mean Years of 11.3 10.7 11.8 12 Schooling (Adults) Recent Economic Experience Average GDP 3.79 3.77 4.04 4.09 Growth (2002-2012) Average Inflation 9.96 8.97 2.78 4.11 (2002-2012) GINI Index 24.8 28.2 33.2 32.2 (2010) Proportion Living .241 .222 .106 .175 Under National Poverty Line Sources: World Values Survey 2010-2014, The World Bank
Levels of education and urbanization are generally similar across all four countries.
However, Ukraine and Romania have significantly lower GDP per capita than Estonia and
Poland. This does not offer strong evidence that modernization is responsible for authoritarian
24 support in this region but we certainly can’t rule out the idea that relative levels of wealth have some effect.
Ukraine is more distinct than Romania for almost all of the variables except for GDP growth, which is only slightly larger than growth in Romania. Although this data does not allow any definitive conclusions, there is certainly some preliminary evidence that economic difficulty may drive an embrace of authoritarianism in Central and Eastern Europe. The degree to which these differences are mediated through citizens’ perceptions of the economy will be explored in the next section.
Subjective Measures
The variables used above capture the economic situation of the countries as a whole but sometimes these numbers can be deceiving or ignore the real day-to-day experience of individuals in the country. For this reason, many authors investigating the issue of regime support have opted to use survey data to determine each individual’s personal experience with the economy rather than using broad macro-level data. 20 This approach assumes that changing regime preferences will be mediated through an individual’s perception of the economic situation rather than the official numbers. The following tables investigate several subjective, personal measures of economic growth. For each region I will list Kendall’s tau-b correlations between authoritarian support and economic factors. The survey questions used for each region can be found in Appendix B.
I have also included the country-specific data for Lesotho and Guatemala in addition to data from the region as a whole. This will help explore the possibility, suggested above, that extremely poor economic conditions could cause a bump in authoritarian support even though
20 Evans and Whitefield; Richard Rose et al., Democracy and it’s Alternatives , The John Hopkins University Press: Baltimore (1998).
25 middling conditions do not have the same effect. The findings are generally consistent with those for the objective variables, showing a limited effect of economic variables that differs from region to region. The analysis of exceptional cases appears to reject the idea that notably poor conditions experienced in Guatemala, Lesotho and Ukraine have a greater effect on authoritarian support.
Latin America
Attempting to capture similar variables to those used above, I have included both short- term and long-term measures of perceptions of the economic situation in Latin America. I look at perceptions of survey-takers’ own as well as their countries’ economic situations. Also investigated are perceptions of fairness of income distribution, the government’s ability to handle poverty, and education levels. The correlations of these variables with various measures of regime support are presented below in Table 2.4.
The results show that, while many elements of an individual’s perception of their own and their country’s economic situation are related to regime support (and generally in the predicted directions), the effects are small. Citizens who view the country’s economic situation poorly are more likely to prefer authoritarianism, accept military rule and believe that democracy is not always the best form of rule. Among generally weak predictors of authoritarian support, the most significant appear to be levels of education and perceptions that the government cannot solve the problem of poverty. Again, these relationships are notably small but they are most powerful for explaining differences in the Churchillian notion that “democracy may have problems but it is the best system of government.” Economically frustrated Latin Americans seem comparatively reluctant to endorse more concrete forms of authoritarian rule.
26 Interestingly, perceptions of unfair distribution of income are not connected with all measures of support for authoritarianism. This may help rule out the idea that the poor, dissatisfied with an unequal system, are clamoring for authoritarian rule. On the contrary, the case of Guatemala may offer some evidence that it is actually the wealthy that tend to find authoritarianism suitable. Lower levels of education are significantly correlated with a decreasing preference for authoritarianism in Guatemala, the reverse of what the modernization literature might predict. However, if we consider the fact that the poor are much more likely to lack education in Guatemala, we may find an explanation to this apparent contradiction in another body of work. Carles Boix has argued that for the “democratic game” to work in a given country, it is crucial that there be some degree of equality of conditions. Absent this level playing ground, the “incentives to cheat become irresistible.” 21 This means that those who benefit most from inequality might be open to authoritarian forms of rule that could prolong their position of strength. It seems entirely possible that this is operating in Guatemala (a highly unequal country), but the degree of inequality’s overarching relevance for authoritarian support in the region as a whole remains unclear. Additionally, I find evidence in Chapter 6 that the division in authoritarian support in Guatemala is more clearly tied to ethnicity than class.
21 Carles Boix, “The Roots of Democracy,” Policy Review (February & March 2006).
27 Table 2.4: Subjective Economic Assessments and Regime Support in Latin America Authoritarianism Support Disagree that Authoritarianism can be preferable military rule democracy is can be better than preferable alternatives (Guatemala ) Country’s .035** .036** .136** .046 Economic Situation Bad Personal/Family .028** .013 .100** .050 Economic Situation Bad Negative Change .020** .014* .086** .007 in Personal/Family Economic Situation Income -.012 -.020** .062** .038 Distribution Unfair Government Can’t .092** .046** .140** .027 Solve the Problem of Poverty Lower level of .056** .067** .066** -.110** Education Attained * significant at .05 level ** significant at .01 level
Perhaps the most surprising finding here is the fact that no subjective measure of the economic situation helps explain authoritarian support in Guatemala. While it is important to point out that the much larger number of cases for the entire region would be affecting statistical significance, the correlations are quite low as well in Guatemala. Only education (an aspect of modernization) seems to be related to levels of regime support among the population. This seems to run contrary to the idea, suggested above, that the extremely poor economic conditions in Guatemala might make the democratic government’s economic performance all the more important for garnering legitimacy. Instead, we see that this kind of performance is irrelevant to regime support in the country.
28 Africa
In Africa, a slightly different set of economic variables was used based on the available survey data. Similar to Latin America, this analysis includes perceptions of both the country’s and one’s personal economic situation. Also used is a measure of how the survey-taker feels their economic situation stacks up to that of other citizens. To capture poverty, I have included a question asking how often the respondent’s family goes without food in a 12-month period. This is different from the question used in Latin America that captures feelings about the government’s capability to alleviate poverty rather than the respondent’s own situation. Finally, to gauge the effect of modernization I have employed levels of education and a simple rural/urban variable for where the survey took place. Again, the specific measures used below can be found in Appendix B.
Consistent with the results from objective economic measures, there seems to be less of an impact of economic assessment on regime support in Africa compared to Latin America.
Most economic measures are statistically significant for explaining some (but not all) measures of authoritarian support. However, their effect is very small. Only poverty and education levels consistently explain regime support and these too have limited impacts.
In Lesotho, short-term economic evaluations as well as impoverished conditions do not help explain the high levels of support for authoritarianism in the country. Modernization indicators, however, do seem to impact regime legitimacy. Levels of education and the rural/urban measure are both statistically significant with correlation coefficients comparable to the highest ones found in the region as a whole.
29 Table 2.5: Subjective Economic Assessments and Regime Support in Africa Rejection of Approval of Approval of Rejection of Authoritarianism One-Party Rule Military Rule Authoritarianism (Lesotho ) Country’s .162 .054** .046** -.046 Economic Situation Bad Personal -.003 .029** .049** -.009 Economic Situation Bad Personal .019** .000 .022** -.014 Economic Situation Better than other citizens Family Often .024** .037** .020** -.051 Doesn’t Have Enough Food Level of .013* -.114** -.080** .126** Education Rural Area -.005 .081** .025** -.085** * significant at .05 level ** significant at .01 level
The data here suggests that we accept some limited impact of economic factors on authoritarian support in Africa. However, this effect is very small, especially compared with the other regions under observation. Only education and poverty help explain all three measures of authoritarian support included in this analysis, albeit with limited predictive power.
Central and Eastern Europe
To measure the effect of the economy on authoritarian support in Central and Eastern
Europe I employed questions included in the World Values Survey. To obtain an idea of overall contentment with living conditions, a measure of satisfaction with one’s personal financial situation is included below. Poverty is measured by the respondent’s answer to how often they
30 or their family go without food. Also included are modernization-related measures of education and urban or rural residence. The relationships of these economic variables with regime support are presented in Table 2.6.
In this region, certain economic variables seem to have greater explanatory power than in
Africa or Latin America, but their effect is still limited. Higher levels of education are consistently associated with preferences against authoritarianism and for democracy. A more educated person is notably more likely to reject military rule in favor of democracy, although the effect on rejection of strongman rule is more limited. Poverty as well is consistently significant and in the direction predicted, although to varying degrees of strength. It is most powerful as an explanation for support for an undemocratic strong leader. Some impoverished Central and
Eastern Europeans demonstrate a willingness to embrace a strongman-style form of rule that could perhaps bring the solutions that democracy has failed to deliver. Both satisfaction with one’s personal financial situation and the size of one’s town are statistically significant but with low correlation coefficients and not always in the predicted direction.
Ukraine generally shows similar trends to the region as a whole. Poverty and low education seem to drive rejection of democracy to a limited degree. Dissatisfaction with personal finances does not appear related to authoritarian support and those living in more rural areas are only slightly more likely to reject democracy. The economic dimension of authoritarian support in Ukraine will be explored further in Chapter 6.
31 Table 2.6: Subjective Economic Assessments and Regime Support in CEE Rejection of Rejection of Rejection of Rejection of Strongman Rule Military Rule Democracy Democracy (Ukraine ) Satisfaction With .053** -.048** -.059** -.026 Personal Financial Situation Family Always .155** .090** -.046** -.077** Has Enough Food Education Level .051** .138** -.106** -.086** Size of Town -.028** .058** -.055** -.045* * significant at .05 level ** significant at .01 level
Evidence from all three regions suggests that, at the individual level, economic conditions and perceptions are not irrelevant to regime commitment. Economic downturn and poverty appear to play a minor role in the formation of regime preferences, with poor conditions driving some to consider non-democratic alternatives like military or strongman rule. Still, consistent with the results from analysis of macro-level economic indicators, the effect of economics is very small and varies across regions and even countries. It is also clear that economic factors do not play a significantly more decisive role in the cases identified with notably poor economic conditions and high popular authoritarian support.
Conclusions
The evidence on whether economic conditions affect authoritarian support is thoroughly mixed and there is significant regional variation in terms of what factors matter the most. In
Latin America, incomplete modernization seems to play a minor role in driving authoritarian support. Countries with a larger proportion of the population living in rural areas have a tendency toward higher mass support for autocracy. The same can be said of lower levels of
32 education at the individual level. The record is mixed on whether short-term considerations matter for authoritarian support in the region, but if they do, their effect is weak.
Country’s that are more unequal tend to have higher popular preferences for authoritarian rule in Latin America. However, zooming in to the individual level to test the effect of perceptions of inequality on authoritarian support leaves us with more questions than answers.
Perceptions of a lack of income equality do not appear to be responsible for driving Latin
Americans to authoritarian alternatives across the board. The impact of an unequal society on authoritarian support, observed at the country level, may be caused by some other, uncovered factor not immediately connected in the minds of citizens.
African preferences for authoritarian rule seem to be highly resistant to varying economic conditions both short-term and long-term. Country-level analysis revealed no important trends while survey data showed very small associations between some economic variables and measures of regime support. Low education and poverty are the only economic indicators that seem to drive authoritarian support, and their effects are small.
In Central and Eastern Europe, the countries with higher authoritarian support had lower average GDP growth, higher inflation and higher levels of poverty. Survey data confirms the effect of poverty and reveals that lower education may be driving regime support to some degree.
It is worth keeping in mind that the small number of cases in this region may be behind the stark differences observed at the regional level.
Some trends can be observed across all three regions. The effects of short-term evaluations of the economic situation tend to have only a very small effect on authoritarian support. This suggests that economic downturn is unlikely to bring calls for the overthrow of newly democratic governments. What seems to matter more are longer-term economic trends.
33 In particular, high levels of poverty and lower education tend to be found where there is higher authoritarian support. The effect of education might be explained by Russell Dalton’s theory that educated individuals will be better able to depend on themselves rather than elites for forming political preferences. 22 The flip side of that would be that less educated individuals are more likely to rely on often-deceptive elites as reference points for determining their preferred regime type. This could potentially lead some to support charismatic, authoritarian populists.
Under, conditions of extreme poverty, citizens could be expected to simply look for a way out rather than putting their faith into slow and weighty democratic institutions.
Additionally, it is highly likely that poverty and low education are often found in the same places and each may reinforce the impact of the other.
22 Russell Dalton, Citizen Politics: Public Opinion and Political Parties in Advanced Western Democracies (Chatham: Chatham House, 1996), 21.
34 Appendix B:
Latin America: Economic Variables Economic Variables Variable Survey Question Ordered Responses Country’s In general, how would you describe the Very good, Good, About Economic country’s present economic situation? average, Bad, Very bad Situation Bad Would you say it is...? Personal/Family In general, how would you describe your Very good, Good, About Economic present economic situation and that of your average, Bad, Very bad Situation Bad family? Would you say it is..? Negative Change Do you consider your economic situation Much better, A little better, in and that of your family to be much better, The same, A little worse, Personal/Family a little better, about the same, a little worse Much worse Economic or much worse than 12 months ago? Situation Income How fair do you think income distribution Very fair, Fair, Unfair, Very Distribution is in (country)? unfair Unfair Government To what extent do you think the state can All the problem, A large part Can’t Solve the solve...the problem of poverty? of the problem Problem of Poverty Level of What level of education do you have? 0-17 years Education What was the last year you completed? Attained Regime Support Variable Survey Question Ordered Responses Authoritarianism With which of the following statements do Democracy is preferable to can be preferable you agree most? any other kind of government; Under some circumstances, an authoritarian government can be preferable to a democratic one Reject Military Would you support a military government I would support a military Rule in replacement of a democratic one if government in replacement of things get very bad? Or would you not a democratic government if support a military government under any things get very bad; I would circumstance? not support a military government under any circumstance Disagree that Do you strongly agree (1), agree (2), Strongly agree, Agree, Democracy may disagree (3) or strongly disagree (4) with Disagree, Strongly disagree have Problems the following statements? ... Democracy But is Still Better may have problems but it is the best
35 Than Alternatives system of government.
Africa: Economic Variables Economic Variables Variable Survey Question Ordered Responses Country’s Economic Situation In general, how would you Very bad, Fairly bad, Neither Bad describe... the present good nor bad, Fairly good, economic condition of this Very good country? Personal Economic Situation In general, how would you Very bad, Fairly bad, Neither Bad describe... your own present good nor bad, Fairly good, living conditions? Very good Personal Economic Situation In general, how do you rate Much worse, Worse, Same, Better than other citizens your living conditions Better, Much better compared to those of other Basotho? Family Often Doesn’t Have Over the past year, how often, Never, Just once or twice, Enough Food if ever, have you or anyone in Several times, Many times, your family gone Always without...enough food to eat? Level of Education What is the highest level of No formal schooling; Informal education you have schooling only (including completed? Koranic schooling); Some primary schooling; Primary school completed; Some secondary school / high school; Secondary school / high school completed; Post- secondary qualifications, other than university e.g. a diploma or degree from a technikon or college; Some university; University completed; Post- graduate Rural Area [Circle One] Urban, Rural Regime Support Variable Survey Question Ordered Responses Rejection of Authoritarianism Which of these three In some circumstances, a non- statements is closest to your democratic government can be own opinion? preferable; Democracy is preferable to any other kind of government Approval of One-Party Rule There are many ways to Strongly disapprove, govern a country. Would you disapprove, Neither approve disapprove or approve of the nor disapprove, Approve, following alternatives? ... Strongly Approve
36 Only one political party is allowed to stand for election and hold office. Approval of Military Rule There are many ways to Strongly disapprove, govern a country. Would you disapprove, Neither approve disapprove or approve of the nor disapprove, Approve, following alternatives? ... The Strongly Approve army comes in to govern the country.
Central and Eastern Europe: Economic Variables Economic Variables Variable Survey Question Ordered Responses Satisfaction With Personal How satisfied are you with the 1 (completely dissatisfied) - Financial Situation financial situation of your 10 (completely satisfied) household? Family Always Has Enough In the last 12 month, how Often, Sometimes, Rarely, Food often have you or your Never family...gone without enough food to eat? Education Level What is the highest No formal education; educational level that you Incomplete primary school; have attained? Complete primary school; Incomplete secondary school: technical/vocational type; Complete secondary school: technical/vocational type; Incomplete secondary: university-preparatory type; Complete secondary: university-preparatory type; Some university-level education, without degree; University-level education, with degree Size of Town (Code size of town ) Under 2,000; 2,000 - 5,000; 5- 10,000; 10 - 20,000; 20 - 50,000; 50 - 100,000; 100 - 500,000; 500,000 and more Regime Support Rejection of Strongman Rule I'm going to describe various Very Good, Fairly Good, types of political systems and Fairly Bad, Very Bad ask what you think about each as a way of governing this country...Having a strong leader who does not have to
37 bother with parliament and elections Rejection of Military Rule I'm going to describe various Very Good, Fairly Good, types of political systems and Fairly Bad, Very Bad ask what you think about each as a way of governing this country... Having the army rule Rejection of Democracy I'm going to describe various Very Good, Fairly Good, types of political systems and Fairly Bad, Very Bad ask what you think about each as a way of governing this country... Having a democratic political system
38 Chapter 3: Politics and Regime Preference A second body of work looking at regime legitimacy argues that rather than delivering on
economic growth, citizens are more inclined to support democracy if their country “delivers on
its promise of freedom and democracy.” 23 This line of thinking suggests that citizens can and
will separate their country’s economic experience from its political experience and avoid turning
against the regime simply because it has not brought the country prosperity. This chapter will
proceed on the basis of this literature on democratic support, mainly because it is the only
literature available and the two variables (democratic and authoritarian support), although
perhaps not exact opposites, are clearly related.
In 1995, Geoffrey Evans and Stephen Whitefield conducted one of the first systematic,
quantitative investigations into the drivers of democratic support looking at Central and Eastern
Europe. 24 They find that support for democracy in this region is primarily influenced by
individual evaluations of regime performance and responsiveness. They note a few important
political conditions that weigh heavily on these regime evaluations. Firstly, when constitutional
conflict is central to the political dialogue, it is likely to turn citizens away from the seemingly
unstable regime type. Polarization between congress and the president is cited as a source of
frustration that can decrease democratic support. Evans and Whitefield also argue that party
attachments can “reflect and develop a sense among the public of an input into the democratic
process”. 25 These political variables, taken as a whole, are much more influential in their statistical analysis than economic factors.
23 Larry Diamond, Developing Democracy: Toward Consolidation , The John Hopkins University Press (Baltimore: 1999), 192. 24 Geoffrey Evans and Stephen Whitefield, “The Politics and Economics of Democratic Commitment: Support for Democracy in Transition Societies,” British Journal of Political Science , Vol. 25, No. 4 (1995), p. 485-514. 25 Ibid, 499.
39 Analyzing 2001 Latinobarómetro data, Marta Lagos finds that political variables are the
most important determinant of democratic support in Latin America as well.26 Lagos argues that, although extreme economic hardship can influence regime support (an idea that may be at odds with the findings from chapter 2), it is the delivery of certain “political goods” that has the greatest effect on support for democracy. For example, one particularly important event that promotes democratic support occurs when there is a peaceful transition of power, preferably to a marginalized societal group. However, Lagos makes the distinction between democratic and authoritarian support, noting that, in general, a lack of democratic support does not mean a surplus of authoritarian support. Rather, many Latin Americans feel indifferent and distant from the electoral process; they don’t support democracy but also don’t embrace any specific alternative. My findings offer support to this notion.
Another study, carried out by Michael Bratton and Robert Mattes, analyzing three
African countries, also shows that political considerations take precedence over economics. 27
They note, similar to Lagos, that the most influential factor across all cases is to what level the survey taker believes the country has delivered on a “basket of political goods.” These include equality before the law, individual rights, and free and fair elections. The provision of these
“goods,” as well as an interest in politics and a general satisfaction with the government’s performance are the best conditions for democratic support.
There are many reasons to believe that the political factors pointed out here will be relevant not only to democratic legitimacy but also popular support for authoritarianism. If general perceptions hold that democracy is not providing freedom and choice as it purports to do,
26 Marta Lagos Cruz-Coke, “A Road with No Return?” Journal of Democracy, 14 (2) (2003), 163-173. 27 Michael Bratton and Robert Mattes, “Support for Democracy in Africa: Intrinsic or Instrumental?,” British Journal of Political Science , 31(3) (2001), 447-474.
40 citizens might be more open to different constructions of political power. Similarly, if democratic institutions are seen as unresponsive, the siren call of a populist leader promising real change could be tempting. The findings below suggest that this is true to a very limited degree in Latin America and Central and Eastern Europe. In Africa, political evaluations offer very little to explain authoritarian support.
Turning to two specific political factors that may be particularly relevant to authoritarian support, I look at corruption and historical repression. High levels of corruption might be expected to undermine regime support. 28 Mitchell Seligson has found that corruption victimization erodes regime legitimacy in Latin America. 29 Corruption may be seen by the population to be fully unproductive rent-seeking that often appears to be permitted by the state.
These kinds of flawed democratic systems could produce a disenchanted population that is willing to give up some freedom in order to bring an end to pervasive corruption. Kurt Weyland has suggested that corruption may have even increased in Latin America under democratic institutions. 30 The diffuse decision making structure of democracy means that their are more points in the policy making process where bribes can be demanded in exchange for consent. In short, there are more palms to grease. Surely there is also more press coverage of high profile corruption cases that authoritarian regimes would have been more likely to successfully suppress.
If democracy and corruption are perceived as growing together, pervasive corruption might give the population good reason to seek an alternative means of governance. My analysis indicates
28 See: Yu-tzung Chang, Yun-han Chu, Chong-Min Park, “Authoritarian Nostalgia in Asia,” Journal of Democracy , Vol. 18, No. 3 (2007) 74. 29 Mitchell A. Seligson, “The Impact of Corruption on Regime Legitimacy: A Comparative Study of Four Latin American Countries,” The Journal of Politics , 64(2) (2002), 408-433. 30 Kurt Weyland, “The Politics of Corruption in Latin America,” Journal of Democracy , 9(2) (1998), 108.
41 that corruption does contribute to authoritarian support, except in Africa where politics, like economics, appear fairly irrelevant for explaining regime legitimacy.
As a final variable, I will investigate levels of political repression to see if a history of regime violence might dissuade support for future authoritarian regimes. It was Rose and his colleagues that introduced the idea of legitimacy being intimately connected with a country’s past political experience. They write: “[t]he alternatives with which democracy competes are other familiar forms of government that have been tried and found wanting.” 31 This means that democracy will tend to be judged in light of the successes and failures of whatever non- democratic regime existed before. Despite the focus on regime history, historical repression levels seem to be left out of their comparative analysis—and the analysis of many who attempt to explain regime legitimacy. We could expect that a high level of violence in a past regime could form a national memory that would dissuade citizens from accepting a similar regime.
Conversely, if political violence is roughly similar or has actually gotten worse under democratic institutions it seems likely that the members of that polity would see the past regime as a viable alternative, or even long for its reestablishment.
This chapter will proceed by first investigating objective, macro-level trends between countries with populations identified as more supportive of authoritarianism and those with lower levels of support. I will then drill down to the individual level using survey data to search for important correlations between more specific political variables and regime support.
Objective Measures
The Freedom House measures of “Freedom in the World” can help provide a glance at the degree to which new democracies have progressed politically. These scores are released
31 Richard Rose, William Mishler, Christian Haerpfer, Democracy and its Alternatives: Understanding Post-Communist Societies , John Hopkins University Press (Baltimore: 1998), 31.
42 annually and rate each country in terms of its political rights and civil liberties. This makes it a useful measure for determining the degree to which each country is able to provide the long list of political goods associated with democracy. Below, I employ each country’s ranking as well as the ten-year change in their scores to account for improvement or decline of political conditions. It is important to note that a lower score implies a more democratic country and scores range from 1-7.
I measure corruption using Transparency International’s Corruption Perceptions Index.
Perceptions are appropriate for this analysis because they will likely have a more direct effect on citizens’ feelings about their own and alternative regimes than actual levels of corruption. For this measure, lower scores indicate a greater prevalence of corruption with a minimum of zero and a maximum of ten.
The Political Terror Scale (PTS) allows us to quantify a country’s history of repression under the previous regime as compared to modern day levels. On this scale, a 1 represents very little political violence while a 5 signifies high levels. The reference point used for determining each country’s “previous regime” was the most recent time when that country’s Polity IV score was negative (meaning it was not a democracy). I took an average of PTS scores starting from one year before that transition and going backwards 5 years. I then took an average of PTS scores in the 3 years leading up to the survey used to determine authoritarian support, a period during which all of the included countries had positive Polity IV scores (2008-2010 in Latin
America and 2005-2007 in Africa). I then subtracted the average during the democratic period from the average during the authoritarian period to find the difference for each country. Central and Eastern Europe are excluded from this analysis because PTS scores were not available.
43 Tables 3.1 and 3.2 present the average of each of these variables for both sets of countries
(more and less supportive of authoritarianism) as well as the difference between them and two- tailed p-values. Again, Guatemala and Lesotho are included for reference as cases with very high authoritarian support. Table 3.3 shows the relevant data for each of the four countries observed in Central and Eastern Europe.
Latin America
In Latin America, popular support for authoritarianism is linked to some political factors
more than others. Table 3.1 shows that countries with higher popular rejection of
authoritarianism are only slightly more democratic based on their freedom house score.
Guatemala, an outlier in authoritarian support, presents an unusually high Freedom House score
(low level of democracy). Unexpectedly, the countries that rejected authoritarianism actually
showed deteriorating political rights and civil liberties, while those that were more willing to
accept authoritarian options improved their democracies. However, these differences do not
reach statistical significance.
Table 3.1: Politics and Regime Support in Latin America Guatemala Countries With Countries With Difference Populations Populations less (p-value) more Willing to Willing to Accept Accept Authoritarianism Authoritarianism (St. Dev.) N=12 (St. Dev.) N=6 Freedom House 7 5.43 5.27 0.16 Ranking 2011 (1.81) (2.72) (0.88) Change in FH rankings 0 -0.14 0.27 -0.41 (2001–2011) (1.22) (1.68) (0.57)
Corruption Perception 2.7 2.84 3.85 -1.01 2011 (0.53) (1.76) (0.12) Change in Political 1.57 0.25 1.75 -1.50 Repression (1.23) (0.40) (0.06) (Positive=Improvement) N=5 N=6 Sources: Latinobarómetro 2011, Freedom House, Transparency International
44 Corruption perceptions are greater in countries with higher popular support for authoritarianism and Guatemala follows this trend. While differences in corruption do not reach statistical significance, there is a much greater difference than for the other measures. On a ten- point scale there is a one-point difference in corruption between the two groups of countries. No country among those with higher popular support for authoritarianism scored above a four on the
Corruption Perceptions Index. Some countries, like Chile and Uruguay, that more readily reject authoritarianism scored seven or higher. While general measures of democracy employed here are inconclusive, there is still a strong possibility that higher perceptions of corruption can cause some citizens to consider authoritarian alternatives.
In Latin America, a region known for its “dirty wars” in response to leftist guerrilla movements, the weight of past repression does seem to have taken its toll. Countries with populations that do not readily reject authoritarianism have, on average, only seen modest improvements (.25 on a scale of 1-5) in the situation of political violence. Meanwhile, the countries where authoritarianism is rejected have seen much less repression under democratic institutions as compared to the previous regime (1.75 on a scale of 1-5). A frightening collective memory of the cruelty of a past regime seems to have left a mark on the populations of these countries. Guatemala is an interesting exception where high authoritarian support has persisted despite a significant decrease in repression. The source of authoritarian support in Guatemala will be investigated in greater detail in Chapter 6.
Another notable exception is Brazil, where the data indicates that repression has been ramped up significantly under democratic institutions yet authoritarian support is not as high as we might expect. Whether an increase in repression has actually occurred is doubtful. To some degree this reflects the difficulty of quantifying levels of repression. My dataset, gathered from
45 PTS scores, consists of averages of two different measures of repression derived from Amnesty
International and the US State Department. In the case of Brazil, the US State Department considered the country to be far less repressive than Amnesty International, resulting in a lower overall score under authoritarian rule. In more recent years, both measures have generally been in agreement. The military dictatorship that ruled Brazil from 1964-1985 did not hesitate to use violence in the form of extrajudicial killings, disappearances and torture. 32 The fact that the
United States at least tacitly—if not fully—backed the 1964 coup could certainly explain the gentler assessment given by the State Department to the military regime. This dark period in the country’s history remains relevant today considering that, as with many Latin American countries, it went unaddressed for many years. The director of Amnesty International in Brazil noted that recent revelations of Brazil’s National Truth Commission amplify “the country’s collective response of ‘Never Again’ to the mass violations of human rights during the authoritarian regime.” 33
The publicity of similar truth commissions in other Latin American countries as well as the collective social memory of events under authoritarian rule appears to be having an important impact on regime preferences. This relationship is shown in Chart 3.1, which demonstrates the strong negative correlation between change in repression and authoritarian support.
32 “Brazil: five decades on, a key step towards truth and justice for dictatorship’s crimes” Amnesty International , December, 10, 2014, accessed April 15, 2015, https://www.amnesty.org/en/articles/news/2014/12/brazil-five-decades-key-step-towards-truth- and-justice-dictatorship-s-crimes/. 33 Ibid.
46 Chart 3.1 : Political Repression vs. Authoritarian Support in Latin America
r = -0.69
Africa
Neither level of democracy, change in level of democracy or perception of corruption has
a significant effect on regime preference in Africa according to this data. The largest difference
between the countries more willing and those less willing to accept authoritarianism occurs in the
case of change in level of democracy. However, in both groups, democracy is improving.
Lesotho is consistent with the other regional averages although its democracy has actually
improved slightly more in the last ten years . This is the opposite of what the literature (which
predicts that improvements will cause higher democratic support) might lead us to expect.
In Africa, the effect of a history of repression is in conclusive, and points in the opposite
direction we woul d expect. Countries with high authoritarian rejection actually saw a greater
47 decrease in political violence, but neither grouping saw very significant improvements at all. It
is perhaps worth noting that, taken as a whole pre-transition repression levels in Latin America
were higher than in Africa (3.46 compared to 2.87). It appears that, among other factors, higher levels of political violence could have formed a more permanent memory that continues to influence citizens’ opinions of the past regime in Latin America, while lower levels of repression have left a smaller impact on African regime support.
Table 2.3: Politics and Regime Support in Africa Lesotho Countries With Countries With Difference Populations Populations less (p-value) more Willing to Willing to Accept Accept Authoritarianism Authoritarianism (St. Dev.) N=9 (St. Dev.) N=10 Freedom House 6 6.11 6 0.11 Ranking 2011 (1.54) (2.54) (0.91) Change in FH rankings -2 -0.56 -1.3 0.74 (2001–2011) (1.01) (2.50) (0.41)
Corruption Perception 3.3 3.31 3.43 -0.12 2011 (0.83) (1.11) (0.79) Change in Political 0.17 0.38 0.11 .27 Repression (0.79) (0.86) (0.55) (Positive=Improvement) N=7 N=8 Sources: Afrobarometer, Freedom House, Transparency International
Central and Eastern Europe
In the Central and Eastern European countries, we can observe most noticeably that the two countries that are more supportive of authoritarian alternatives also have higher Freedom
House scores. This is an interesting finding, although it is difficult to determine its significance with so few cases. There is almost no difference between ten-year changes in Freedom House scores, meaning that the level of development of democracy over time does not appear to be an
48 important factor. All countries became more democratic except for Romania, which regressed slightly.
There is also a higher perception of corruption in the countries with populations that are more supportive of authoritarianism. This is consistent with the idea that disaffection with the failures of the democratic system might lead to disaffection with the idea of it as a whole.
Although the cases are limited, there is some evidence here that politics is related to regime support in Central and Eastern Europe.
Table 3.1: Politics and Regime Support in Four Central and Eastern European Countries Higher Authoritarian Support Lower Authoritarian Support Ukraine Romania Poland Estonia Freedom House 7 4 2 2 Ranking 2011 Change in FH -1 0 -1 -1 rankings (2001– 2011) Corruption 2.3 3.6 5.5 6.4 Perception 2011 Sources: World Values Survey 2010-2014, Freedom House, Transparency International Regional differences persist in this analysis of objective political measures. In CEE, low levels of democracy attained as well as high perceptions of corruption can be observed in countries with more authoritarian support. In Latin America, the same is true of corruption, but levels of democracy do not help predict regime support. Africa continues to evade explanation with only miniscule differences between the two groupings of countries. We now turn to micro- level data to look for confirmation of these results.
Subjective Measures As with economic conditions, measures from international ranking organizations may sometimes fail to capture the reality for citizens in these countries. A one point bump in
Freedom House’s ranking may not be immediately visible to the entire population. With
49 findings as limited as those above, it makes sense to look for other ways to measure the effect of politics on regime preference. Digging in to survey data is one way of getting to the individual level to determine what considerations factor into decisions about regime preference. In the section below, I will look for correlations between various survey questions relating to the political variables from the literature outlined above and several measures of regime support.
The large number of cases for these international surveys may influence the fact that many variables are statistically significant so I will look not only at significance but, more importantly, at the magnitude of the correlations.
Latin America
The data used for Latin America comes from the Latinobarómetro survey carried out in
2011. Many independent variables are used to uncover popular perceptions of the functioning of democracy. To capture these perceptions at the most basic level, I have employed a question asking how satisfied the respondent is with democracy and another asking whether democracy has improved or worsened. Next, to capture the delivery of what some scholars have called a
“basket” of political goods, I built a composite measure that includes questions asking respondents about a number of general conditions that are normally expected of a democratic regime. Among these are; guarantees of social justice, consolidation of political parties, citizen participation, reduced corruption, and increased transparency. Admittedly, these items do not cover all of the so-called goods that democracy is supposed to provide but the available survey data is somewhat limiting.
50 As a proxy for basic electoral efficacy I have included a measure of how likely it is that a
survey-taker’s preferred political sector would make it to power. Also included is a question that
asks how well the government is able to deal with the problem of corruption.
I run each of these factors against several measures of regime support. The first two
measures seek to capture support for authoritarianism based on the percentage who agree that
non-democratic rule can be preferable “under some circumstances” and the portion that approve
of military rule. The third dependent variable captures a lack of democratic support as
measured by whether the survey-taker agrees that democracy is always the best form of
government. The final measure seeks to capture indifference to regime type in order to
determine whether defection from democracy causes support for authoritarianism or simply a
feeling of hopelessness. All of the survey questions used in the tables below are presented in
Appendix C. In general, we can expect that where democratic conditions are seen as lacking,
support for authoritarianism will increase.
To simplify the presentation of this data (since survey questions often use a 1-10 point system or multiple possible answers), the tables below will show bivariate Kendall’s tau-b correlations between each survey question and various measures of regime support. For this method to work, non-answers and answers like “don’t know” have been excluded from this analysis. Table 3.4 has been set up so that positive correlations are consistent with the predictions based on the literature above (meaning that democratic failures drive authoritarian support, decrease democratic support, or increase indifference).
Perhaps the most notable finding from Table 3.4 is the fact that different types of regime support have different sources in Latin America. Negative assessments of the progress of democracy appear critical to explaining low democratic support and indifference. They are
51 somewhat less important, however, for explaining authoritarian support. The data below seems to indicate that poorly functioning democratic institutions cause some individuals to stop supporting the democratic regime and to instead resign themselves to indifference: “for people like me, it doesn’t matter whether we have a democratic or non-democratic regime.” A smaller, but still significant, number of people in this situation are driven to support authoritarian alternatives.
The most powerful drivers of regime support are more general assessments of the democratic system. Dissatisfaction with democracy and perceptions of negative change in the system are associated with authoritarian support and, to a greater degree, indifference and defection from democracy. Failure to deliver on political goods has a very limited effect on support for authoritarianism and indifference but no effect on democratic support. This is somewhat consistent with the findings from objective measures, which saw small but comparable differences in levels of democracy at the country level.
Also important are perceptions of the government’s ability to handle corruption and the efficacy of the system. Those who see corruption as a problem that the government cannot solve are, unsurprisingly, more indifferent and more likely to disapprove of democracy. They are also more likely to embrace authoritarian alternatives. This confirms the tentative finding above that high levels of corruption could drive authoritarian support. Those who feel that their preferred political sector cannot make it to power are similarly supportive of authoritarianism over democracy, or simply indifferent to regime type.
52 Table 3.4: Support for Authoritarianism and Politics in Latin America Agree that Under Support for Disagree that Indifference some military rule democracy is the circumstances an best form of authoritarian government government can be preferable to a democratic one Democracy has .089** .058** .200** .138** worsened Failure to deliver .029** .051** -.007 .045** the political “goods” of democracy
Individual’s .071** .018* .103** .078** party’s ideas can’t make it to power Unsatisfied with .109** .062** .235** .230** how democracy is working
The state cannot .089** .036** .142** .094** solve the problem of corruption * Significant at the .05 level ** Significant at the .01 level
Africa
Thanks to the Globobarometer’s efforts to coordinate these national surveys, many of the questions asked in Latin America were also asked in Africa, with some differences in wording.
For example, rather than asking how democracy is progressing, the Afrobarometer asks how much of a democracy the respondent’s country is “today.” This, along with satisfaction with democracy, are used to capture general feelings of how the system is functioning.
The delivery of political “goods” draws from a line item list of elements of democracy (in this case “freedoms”) including the freedom to “say what you think,” “join any political
53 organization you want,” and to “choose who to vote for without feeling pressured.” The
Afrobarometer also provides an even better measure of democratic efficacy by asking if the survey-taker feels that they could get together with like-minded citizens and change the mind of their national representative. Finally, corruption is investigated in terms of how well the respondent believes the government is handling the issue.
Again I have employed multiple dependent variables including two measures of authoritarian support (support for one-party rule and military rule), one measure of support for
“non-democratic” rule in general, and a measure of indifference to regime type. The complete list of survey questions used in this analysis can be found in Appendix C. Like the
Latinobarometro data, these will be presented using correlations in order to simplify. Again these are set up so that we should see positive correlations if perceptions of democracy are having the expected effects on regime support.
Consistent with the findings from objective variables, the survey data shows almost no impact of political factors on authoritarian support. Only one variable—failure to deliver on political goods—is significant and in the expected direction for all four measures. All of the other variables included indicate a very slight bump in authoritarian support as democracy improves , the reverse of our expectation.
Support for non-democratic rule in general and indifference are somewhat in line with expectations but the effect is quite small. Again, indifference appears to increase as democracy is perceived as being worse. Taken as a whole, there is very little evidence here that perceptions about the political system affect authoritarian support. The only pair that reaches a significant .100 correlation coefficient is the positive effect of failure to deliver the “basket” of political goods on indifference.
54 Table 3.5: Support for Authoritarianism and Politics in Africa Approve of one- Approve of Preference for Indifference party rule military rule non-democratic government Country is not -.061** -.021** .036** .049** governed very democratically Failure to deliver .020** .052** .078** .100** political goods of democracy Unlikely that you -.034** -.010 -.037** .029** could influence national representative Unsatisfied with -.062** -.004 .077** .057** democracy Government not -.083** -.039** -.004 -.005 handling corruption well Source: Afrobarometer
Central and Eastern Europe
The World Values Survey offers a different set of political variables related to the state of democracy. Questions asking how democratically the country is being governed as well as whether human rights are respected capture general democratic conditions. These are supplemented by perceptions of the fairness of elections and whether elections are bought by the rich. The final question used below looks at the degree to which the state is able to provide safety and security to its citizens. This is a condition that was not investigated in the other regions but we should expect authoritarian support where democratic governments are failing to keep their people safe. As with Latin American and Africa, positive correlations indicate that democratic deficits are driving authoritarian support or indifference.
The effect of political factors on military rule and rejection of democracy is weak and not always in the predicted direction. A perception of lack of respect for human rights, lack of security and unfair elections all diminish the importance of democracy in the eyes of citizens, but
55 only by a small amount. This indicates that poorly functioning democratic institutions are responsible for some indifference, although not at the levels found in Latin America.
Table 8: Politics and Regime Support in Central and Eastern Europe Approve of Approve of Reject Democracy not strongman rule military rule democracy important How -.092** .038** -.046** -.078** democratically is the country being governed today No respect for .230** .050** .027* .040** human rights in country Don’t feel secure .036** -.026* .023 .050** in your neighborhood Elections are not .227** .027* .055** .078** fair Elections are .202** .027* .007 -.003 bought buy the rich Source: World Values Survey 2010-2014 In Central and Eastern Europe support for a strong leader who does not have to bother with parliament or elections is the most popular authoritarian alternative. According to this data, it is also the measure of authoritarian support that is best explained by the political context of the country and its progression toward democracy. Poorly respected human rights, unfair elections and a perception that elections are bought by the rich are all strongly and significantly associated with support for strongman rule. A perception of failing democratic institutions is causing a number of citizens to desire a strong central leader who can fix the broken system. Furthermore, if citizens feel that elections are not functioning and giving them a voice, perhaps a benevolent strongman ruler would be better than a corrupt politician. The effect of political factors on support for strongman rule is the most important evidence yet that politics does, in some cases, affect regime support.
56 The oddball case of Ukraine will be explored in a later chapter but it is worth noting that these general trends do not hold in that country. Political variables are largely irrelevant to the high levels of authoritarian support in Ukraine. Dropping Ukraine out of this analysis actually increases many of the correlation coefficients presented below for approval of strongman rule.
Conclusions Much of the literature on democratic legitimacy has centered on the effect of political variables. The main assertion has been that a successful delivery of democratic institutions will build support for those institutions. The preceding investigation has taken these predictions and attempted to apply them to support for authoritarianism (presumably the opposite side of the coin). This chapter confirms that, as with economic variables, not all regions are created equal.
There are significant differences in the relative importance of very similar political variables across Africa, Latin America and Central and Eastern Europe. However, some generalizations are possible.
Consistent with the findings of Marta Lagos, negative perceptions of the workings of democracy, as well as dissatisfaction with it, are linked to increasing indifference to regime type.
This indifference is most pronounced in Latin America and present to a lesser degree in Africa and CEE. In Latin America, weak democracy does seem to be driving a small number of people to authoritarian support as well. Only in CEE do we find political variables clearly and strongly linked to support for authoritarianism, in the form of strongman rule.
In Latin America and CEE, both objective and subjective measures confirm the importance of corruption as a small part of the explanation for authoritarian support. Corrupt politics in Latin America drive citizens in nearly equal measures to authoritarian support and indifference. While this is not the strongest finding, it is notably consistent across various modes
57 of measurement. Similarly, in CEE, perceptions that money influences elections are a strong and significant predictor of support for strongman rule.
Although democratic deficiencies cause a small amount of indifference in Africa, politics as a whole do not seem particularly important for explaining regime support in general. The political variables employed for Africa offer no help in explaining authoritarian support in particular.
An important finding in this chapter is related to regime history. It seems that a historical experience of highly repressive authoritarian rule in Latin America leads some individuals to say
“never again.” The weight of the past appears critical to understanding present assessments of different regime types. This suggests that preferences for authoritarian rule are determined with an eye to past experience, even if it is only a distant memory.
The preceding analysis demonstrates that political factors offer more insights into authoritarian support than economic factors (with the possible exception of Africa). However, democratic deficiencies tend to be much better predictors of indifference than support for autocracy. This may make sense, considering that if one is disappointed with a lack of freedom and voice, an authoritarian alternative does not seem like a likely candidate to provide it. The example of the importance of corruption for explaining regime support in Latin America may indicate that the formulation of authoritarian support is carried out with reference to past experience with similar regimes. Some Latin Americans might see democratic regimes as more corrupt than their authoritarian predecessors and see a return to previous institutions as a possible solution to this critical problem. The next chapter will continue to explore the importance of previous experience with autocracy for determining authoritarian support in Latin America.
58 To conclude, it seems that democratic deficiencies alone are not likely to turn large portions of the population toward supporting illiberal regime types. However, the small portion of individuals that do seem to make this shift, combined with the larger portion who become increasingly indifferent, could still provide an apt environment for an authoritarian regime to rise to power largely unopposed. For this reason, political variables should be considered an important piece of the puzzle of authoritarian support, but there is clearly still more at play.
59 Appendix C Latin America
Independent Variables Title Survey Question Responses Democracy has worsened Do you think that democracy Improved, Remained the in (country) has...? same, Worsened Failure to deliver the political What does democracy still 0 (none mentioned) – 5 (all “goods” of democracy need to do in your country?... mentioned) Consolidate political parties, Guarantee social justice, Increase citizen participation, Reduce corruption, Increase transparency of the state Individual’s party’s ideas Do you believe that the Has the same opportunities, can’t make it to power political sector which you Does not have the same support has the same opportunities opportunities to get into power as others? Or you don’t think it doesn’t have the same opportunities? Unsatisfied with how In general, would you say you Very satisfied, Quite satisfied, democracy is working are very satisfied, quite Not very satisfied, Not at all satisfied, not very satisfied or satisfied not at all satisfied with the working of the democracy in (country)? The state cannot solve the To what extent do you think All of the problem, A large problem of corruption the state can solve... the part of the problem, A small problem of corruption? part of the problem, The state cannot solve the problem Dependent Variables Title Survey Question Responses Agree that Under some With which of the following Democracy is preferable to circumstances an authoritarian statements do you any other kind of government; government can be preferable agree most? Under some circumstances, an to a democratic one authoritarian government can be preferable to a democratic one Support for military rule Would you support a military I would support a military government in replacement of government in replacement of a democratic one if things get a democratic government if very bad? Or would you not things get very bad; I would support a military government not support a military under any circumstance? government under any circumstance
60 Disagree that democracy is the Do you strongly agree (1), strongly agree, agree, best form of government agree (2), disagree (3) or disagree, strongly disagree strongly disagree (4) with the following statements? ... Democracy may have problems but it is the best system of government. Indifference With which of the following “For people like me, it doesn’t statements do you matter whether we have a agree most? democratic or non-democratic regime” or “Don’t Know”; any other response
Africa
Independent Variables Title Survey Question Responses Country is not governed very In your opinion how much of A full democracy; A democratically a democracy is Lesotho today? democracy, but with minor problems; A democracy with major problems; Not a democracy Failure to deliver political In this country, how free are 0 (not at all free/not very free goods of democracy you: ... To say what you think, for all) - 3 (somewhat To join any political free/completely free for all) organization you want, To choose who to vote for without feeling pressured Unlikely that you could In your opinion, how likely is Not at all likely, Not very influence national it that you could get together likely, Somewhat likely, Very representative with others and make: ... Your likely representative to the National Assembly listen to your concerns about a matter of importance to the community? Unsatisfied with democracy Overall, how satisfied are you Very satisfied, Fairly satisfied, with the way democracy Not very satisfied, Not at all works in Lesotho? Are you: satisfied Government not handling How well or badly would you Very badly, Fairly badly, corruption well say the current government is Fairly well, Very well handling the following matters, or haven’t you heard enough to say? Dependent Variables Title Survey Question Responses Approve of one-party rule There are many ways to Strongly disapprove,
61 govern a country. Would you Disapprove, Neither approve disapprove or approve of the or disapprove, Approve, following alternatives? ... Strongly approve Only one political party is allowed to stand for election and hold office. Approve of military rule There are many ways to Strongly disapprove, govern a country. Would you Disapprove, Neither approve disapprove or approve of the or disapprove, Approve, following alternatives? ... The Strongly approve army comes in to govern the country. Preference for non-democratic Which of these three Democracy is preferable to government statements is closest to your any other kind of government; own opinion? In some circumstances, a non- democratic government can be preferable. Indifference Which of these three For someone like me, it statements is closest to your doesn’t matter what kind of own opinion? government we have; any other response
Central and Eastern Europe
Independent Variables Title Survey Question Responses How democratically is the And how democratically is 1 (not at all democratic) – 10 country being governed today this country being governed (completely democratic) today? No respect for human rights in How much respect is there for A great deal of respect for country individual human rights individual human rights, nowadays in this country? Do Fairly much respect, Not you feel there is... much respect, No respect at all Don’t feel secure in your Could you tell me how secure Very secure, Quite secure, Not neighborhood do you feel these days in your very secure, Not secure at all neighborhood ? Elections are not fair In your view, how often do the Very often, Fairly often, Not following things occur in this often, Not at all often country’s elections? ... Election officials are fair Elections are bought buy the In your view, how often do the Very often, Fairly often, Not rich following things occur in this often, Not at all often country’s elections? ... Rich people buy elections Dependent Variables
62 Title Survey Question Responses Approve of strongman rule I'm going to describe various Very good, Fairly good, Fairly types of political systems and bad, Very bad ask what you think about each as a way of governing this country...Having a strong leader who does not have to bother with parliament and elections. Approve of military rule I'm going to describe various Very good, Fairly good, Fairly types of political systems and bad, Very bad ask what you think about each as a way of governing this country...Having the army rule Reject Democracy I'm going to describe various Very good, Fairly good, Fairly types of political systems and bad, Very bad ask what you think about each as a way of governing this country...Having a democratic political system Democracy not important How important is it for you to 1 (not at all important) – 10 live in a country that is (absolutely important) governed democratically?
63 Chapter 4: Culture, History, and Authoritarian Support The scholarly debate around democratic and regime legitimacy has tended to center on
the relative importance of economic and political factors. However, the data from Chapters 2
and 3 suggest that these two sets of explanatory variables cannot explain the entire picture alone.
Some authors have—both implicitly and explicitly—pointed out cultural and historical variables
that may influence the regime preferences of specific populations. This chapter does not attempt
a complete, in-depth investigation into the impact of the very complex and rich concepts of
history and culture. Rather, what follows will investigate several specific variables brought up in
or related to the literature on democratic and authoritarian legitimacy. First, I will use multiple
measures to investigate the claim that certain countries may have cultural traits that are hostile to
democracy. Second, I will look for connections between levels of trust and authoritarian support.
Culture
In The Third Wave , Samuel Huntington sums up the cultural argument: “A profoundly
antidemocratic culture would impede the spread of democratic norms in the society, deny
legitimacy to democratic institutions , and thus greatly complicate if not prevent the emergence
and effective functioning of these institutions” (my emphasis). 34 The legitimacy argument here
is that where there is no Western culture and/or where there is some other culture hostile to
democratic norms and behavior, support for democracy will be slow to develop. We may
assume that at least some of these non-democrats would have some other illiberal regime in mind
as a preferable option. While he does not fully back this argument, Huntington does point to
Confucian culture as solidly anti-democratic and Islamic culture as at least problematic for the
34 Samuel P. Huntington, The Third Wave: Democratization in the Late Twentieth Century (Norman: University of Oklahoma Press, 1991), 298.
64 development of a democratic culture. In fact, some studies looking at Asia have confirmed this idea of a set of particular “asian values” that can undermine democratic legitimacy and lend legitimacy to autocrats. 35
The focus of this investigation is on three regions not mentioned by Huntington: Latin
America, Africa, and Central and Eastern Europe. In Africa and Latin America, it would be difficult to pinpoint a specific cultural trait that is particularly hostile to democracy. However, in the case of Latin America, some early observers pointed to authoritarian and hierarchical cultural traits that could undermine democratic support. 36 In Africa, far from discovering some united cultural identity, an investigation carried out by Michael Bratton, Robert Mattes and E. Gyimah-
Boadi concludes that it is “a culturally contested continent.” In his discussion of culture,
Huntington does mention the possibility of a cultural impediment to democracy in the former
Soviet Union. He suggests that countries like Ukraine, Romania and Estonia were part of a unique cultural experience that was more influenced by the Czarist and Ottoman Empires,
Othodoxy, and Islam. 37 Rose, Mishler and Haerpfer investigated this idea by measuring the impact of religion (secular vs. religious), remoteness (rural vs. urban) and the importance of nationality to an individual’s regime preferences in this region. They found only a very weak connection between these cultural variables and support for authoritarian alternative regimes. 38
35 Chong-Min Park, Doh Chull Shin, “Do Asian Values Deter Popular Support For Democracy? The Case of South Korea,” Asian Barometer Working Paper Series (2004). 36 Howard Wiarda, ed., Politics and Social Change in Latin America: The Distinct Tradition (Amherst: University of Massachusetts Press, 1974). 37 Huntington, 299. 38 Richard Rose, William Mishler, Christian Haerpfer, Democracy and its Alternatives: Understanding Post-Communist Societies , John Hopkins University Press (Baltimore: 1998), 138.
65 Despite the mixed record in the literature on the potential impact of culture for authoritarian and democratic support, it may still be valuable to probe these variables in slightly different ways.
One potential way to quantify “Western culture” comes from Ronald Inglehart and
Christian Welzel’s index of emancipative values (EVs), which measure citizens’ democratic orientations in various countries. EVs give “priority to equality over patriarchy, tolerance over conformity, autonomy over authority, and expression over security.” 39 Since this index is based on the World Values Survey, we are limited to exploring only the countries included in the 2011 survey. Further limiting the amount of data points, but in the interest of consistency, I have included only countries that were in the initial lists for the three regions. Still, this data serves as a good starting point for an investigation of culture.
Welzel and Inglehart find a strong correlation between these emancipative values and levels of democracy across the world. They suggest that this is due to a high priority placed on democratic values that translates into mass demands for liberal institutions. 40 We have good reason to expect that the inverse would be true as well. Preference for patriarchy, conformity, authority and security should translate into a desire for stable authoritarian rule. The charts below show that this is not the case.
39 Christian Welzel and Ronald F. Inglehart, “Political Culture, Mass Beliefs, and Value Change,” in Democratization , ed. Christian W. Haerpfer et. al. (New York: Oxford University Press, 2009), 129. 40 Ibid.
66 Chart 4.1: Emancipative Values vs. Support for Military Rule
r = -0.16 Chart 4.2: Emancipative Values vs. Support for Strongman Rule
r = 0.04 Source: World Values Survey 2011
67 Charts 4.1 and 4.2 show the relationship between emancipative values and two different measures of authoritarian support across the three regions under investigation. Support for strongman rule was determined by the percentage of respondents in each country that said it was either very good or fairly good to have a “strong leader who does not have to bother with parliament or elections.” The proportion saying that having the army govern was fairly good or very good determined support for military rule. 41 Surprisingly, there is very little relationship at all between both measures of support for authoritarian rule and emancipative values. As an example, Ghana is an outlier due to its extremely low emancipative values, indicating that democratic institutions should not be compatible with the culture. However, there is very little support for authoritarian alternatives in the country. Ghanaians strongly support democracy with
95.6% saying democracy is either a fairly good or a very good regime type for their country. It seems clear that seemingly authoritarian cultural traits are not a guarantee of support for authoritarian alternatives, nor are they an impassible barrier to support for democracy.
To further probe this unexpected finding, I employ a similar variable, found variously worded in several international surveys, that measures individuals’ priorities at the national level.
For example, the Afrobarometer asks respondents to list the first and second most important area of focus from four options: “maintaining order in the nation,” “improving economic conditions for the poor,” “giving people more say in government decisions,” and “protecting peoples right to live freely.” The first two aims are certainly not authoritarian in nature, but an individual that picks the two answers together as their first and second choice may be more predisposed to the temptation of a charismatic autocrat promising economic reform and an end to a violent or
41 WORLD VALUES SURVEY Wave 6 2010-2014 OFFICIAL AGGREGATE v.20141107. World Values Survey Association (www.worldvaluessurvey.org). Aggregate File Producer: Asep/JDS, Madrid SPAIN.
68 dangerous political environment. Table 4.1 examines the level of these kinds of “passive authoritarian values” across three regions in two different sets of countries determined by their levels of authoritarian support. The percentages listed for Africa show the proportion of those surveyed who selected both non-democratic goals. In Latin America, passive authoritarian values were determined by the percentage who chose fighting rising prices and maintaining order rather than giving people more voice in policy decisions and improving freedom of speech. For these measures I had to rely on the World Values Survey, which significantly decreased the number of cases. Additionally, some of the measures included below from Latin America were from survey years other than the 2011 survey used to determine authoritarian support. However, based on other measures contained in the WVS, authoritarian support in these countries has remained fairly steady over the last few years. In order to more closely investigate this region’s selected case, Guatemala, I included data from the same survey question asked several years earlier. The fact that support for a “strong ruler who does not have to bother with parliament and elections” was at 56.4% in that survey indicates that authoritarian support was equally if not more present in that survey year. The same measure of passive authoritarian values used in Latin
America was used for CEE.
The data presented in Table 4.1 demonstrates some significant regional differences. In
Latin America, populations with higher authoritarian support tended to place non-democratic concerns of security and the economy above more liberal goals. Demonstrating exactly the opposite, African countries with higher authoritarian support were more likely to choose democratic goals. This is an interesting and unexpected observation and may reflect differing priorities in the two regions.
69 Table 4.1: Passive Authoritarian Values and Support for Authoritarianism Case Study Countries With Countries With Difference Populations Populations Less (p-value) More Willing to Willing to Accept Accept Authoritarianism Authoritarianism (St. Dev.) (St. Dev.) Proportion Guatemala 28.35% 21.53% 6.82% Choosing Non- (2004) (2.05) (4.34) (0.15) Democratic Goals 31.4% in Latin America Proportion Lesotho 25.98% 35.93% -9.95% Choosing 14.8% (13.03) (11.04) (0.11) Democratic Goals N=9 N=10 in Africa
Countries With Populations More Countries With Populations Less Willing to Accept Authoritarianism Willing to Accept Authoritarianism Proportion Ukraine Romania Poland Estonia Choosing 78.6% 65.8% 60.6% 69.7% Non- Democratic Goals in Central and Eastern Europe Sources: World Values Survey, Afrobarometer
The cases chosen for Latin America and Africa follow regional trends. Guatemalans had extremely high passive authoritarian values in 2004, higher than any other country in the survey.
Conversely, Basotho have the second lowest passive authoritarian values among the countries observed in the region. This demonstrates a difference in the demand for democracy in these regions, but does not help as much for explaining support for authoritarianism. This is especially true in the case of Africa, where low passive authoritarian values (meaning at least some demand for democracy) should, but do not translate into rejection of authoritarian alternatives. Perhaps
Africans see democracy—rather than authoritarianism as we might expect—as a solution to problems of security and the economy. In Latin America, a preference for non-democratic
70 country goals may help explain a small part of the picture. However, the differences between
country groups narrowly miss statistical significance at the .1 level in both regions.
In CEE, there is a much higher demand for non-democratic over democratic goals across
all four countries. These levels are generally about double those found in Latin America and do
not vary greatly among the countries. However, it is interesting to note that Ukraine shows the
highest preference for non-democratic goals. Perhaps the particular importance of stability and
the economy for Ukrainians contributes to a preference for some quicker, more efficient non-
democratic regime. This will be examined further in Chapter 5.
Trust
Another important cultural factor that can facilitate the emergence of democracy, and
may be relevant to these regions, is trust. Marta Lagos has found evidence of a “regional
heritage of distrust” 42 in Latin America that could prove detrimental to the growth of democratic
norms. On the contrary, in Africa, Bratton, Mattes and Gyimah-Boadi found no relationship
between interpersonal trust and support for democracy. However, they did find that more
trusting individuals are actually slightly more likely to support authoritarian alternatives , a
relationship that they attribute to naivety or innocence that can be taken advantage of by
authoritarian rulers. 43
42 Marta Lagos, “Latin America’s Smiling Mask,” Journal of Democracy (1997): 129. 43 Michael Bratton, Robert Mattes, E. Gyimah-Boadi, Public Opinion, Democracy, and Market Reform in Africa (New York: Cambridge University Press, 2005), 197.
71 Table 4.2: Trust and Authoritarian Support Case Study Countries With Countries With Difference Populations Populations Less (p-value) More Willing to Willing to Accept Accept Authoritarianism Authoritarianism (St. Dev.) (St. Dev.) Level of Guatemala 24.04% 21.78% 2.26% Interpersonal 21.3% (8.80) (4.54) (0.55) Trust in Latin N=7 N-11 America Level of Lesotho 46.6% 36.93% 9.67% Interpersonal 46.2% (15.12) (11.47) (0.16) Trust in Africa N=9 N=10
Countries With Populations More Countries With Populations Less Willing to Accept Authoritarianism Willing to Accept Authoritarianism Level of Ukraine Romania Poland Estonia Interpersonal 23.1% 7.7% 22.2% 39% Trust in Central and Eastern Europe
Table 4.2 shows the effect of levels of trust on authoritarian support in general across all
three regions. In Africa, my findings support those of Bratton and his colleagues that lower
levels of trust actually seem to decrease authoritarian support. It may be that, in this case, a
certain level of suspicion can help some individuals avoid being deceived by the alluring
promises of an autocrat. The same trend is visible in Latin America but to a fairly
inconsequential degree. The case in Central and Eastern Europe is similarly inconclusive.
Taken as a whole, the two countries with higher popular support for authoritarianism have lower
levels of trust but this is driven mainly by the extremely low level in Romania. In that country,
only 7.7% of those surveyed trust other citizens. Trust in Ukraine (high authoritarian support) and Poland (low authoritarian support) is roughly the same. Similar to Latin America, this
72 region appears to have very low levels of trust between its citizens, but at these extremes it does not appear to affect regime preference.
Taken as a whole, the cultural explanations explored above offer only tentative, incomplete explanations of variance in regime support. Emancipative values, designed to measure democratic tendencies, are a surprisingly weak predictor of support for illiberal regimes.
In Africa, passive authoritarian values (the proportion of a population that prefer non-democratic goals over democratic goals for their country) do not cause support for authoritarianism. This is important as it suggests that democracy may be seen as capable of handling the basic issues of security and the economy.
Preferences for stability and economic growth over deepening democracy might still have some effect in the other regions, especially at the extremes as demonstrated in Guatemala and
Ukraine. A closer look at Ukraine offered in the next chapter seems to indicate that culture may have a distinct effect in different countries and regions. Finally, trust is not a significant predictor of authoritarian support but it may be somewhat relevant in Africa where lower levels of trust are associated with lower authoritarian support, perhaps due to increased cynicism.
Overall, the evidence that culture determines authoritarian support uniformly across these regions is underwhelming.
Conclusions
This chapter began with the assumption that cultural traits might drive countries to greater authoritarian support. Although this topic could merit another full volume of research, a cursory investigation carried out here suggested that this is not the case. Ronald Inglehart and
Christian Welzel’s emancipative values do not appear related to authoritarian support. Looking
73 more specifically at citizens’ preferences however did reveal some interesting differences. In
Latin America, countries that put a higher priority on economic stability and security tended to have higher levels of authoritarian support. Again, Africa refuses to fit the mold and presented the opposite trends. This seems to indicate that some Africans feel that democracy, security and stability can coexist. However, these non-democratic priorities were only slightly different between the country groups and did not attain statistical significance. This chapter was unable to find much evidence of an effect of culture on support for autocracy in Central and Eastern
Europe.
Overall, culture seems to be a weak predictor of authoritarian support, especially considering the closeness of many of the independent and dependent variables in this analysis.
However, the next chapter, looking specifically at Ukraine, will demonstrate that culture is not totally irrelevant, and might even be central to explaining authoritarian support in the culturally divided region of Central and Eastern Europe.
74 Chapter 5: Authoritarian Support in Ukraine On November 22, 2004, hundreds of thousands of Ukrainians gathered at Independence
Square in Kyiv to protest falsified election results that were set to return the incumbent president,
Viktor Yanukovych, to power. These mobilizations were not confined to the capital and could
be found all across the country. The Orange Revolution, as the protests were called, has been
heralded as a prime example of “people power” where citizens took to the streets to demand
democracy. 44
However, survey data collected by the World Values Survey in Ukraine casts a dark
shadow on this hopeful movement and the brief progress toward democracy that it brought.
Despite fairly high levels of democratic support (only 15% saying democracy is a bad regime
choice), a whopping 71.3% of surveyed Ukrainians approved of “having a strong leader who
does not have to bother with parliament and elections.” That number was 51.8% in 2006 with
nearly 20% of survey-takers responding with “I don’t know.” What explains this majority
preference for what I will term “strongman rule,” even in the context of a seemingly democratic
mass uprising?
This investigation will follow the framework laid out in the previous chapters in order to
solve this puzzle, focusing on the 2011 survey data. 45 The first section will explore economic
factors. I find that economic changes since Ukraine’s transition to democracy are unhelpful for
explaining regime preference. The next section looks at political factors focusing on the current
regime’s delivery of freedom and democracy. These political factors help explain relatively low
levels of support for military rule but they offer little to explain the more impressive levels of
support for strongman rule. Finally, I will explore various cultural traits of the country. I find
44 Adrian Karatnycky, “Ukraine’s Orange Revolution,” Foreign Affairs , 84(2), (2004), 35. 45 This chapter seeks to explain a snapshot in Ukrainian history. It will not take into account more recent, post-2011 events that have dramatically changed the country’s course.
75 that aspects of culture are in large part responsible for the contradictory narrative of regime support in Ukraine.
The Economy
If, as Adam Przeworski predicts 46 , regime support will be generated by how new institutions stack up to popular expectations, Ukraine would stand out as a case where poor economic performance would be likely to have engendered popular preferences for a return to comparatively stable authoritarian rule. Ukraine never fully recovered from its transition to a market economy in the 1990s. Between 1989 and 1999 it experienced the worst economic collapse of all the former-Soviet countries. 47 Today its economy is only ¾ the size it was in
1989 and it has faced continually poor conditions with only a short exception of GDP growth in the early 2000s. Not only did the transition to democracy occur simultaneously with these far- reaching economic reforms, but democracy itself is somewhat implicated in the country’s failure to piece together a working economy. Frequent changing of government has undermined any one administration’s ability to pursue a coherent reform program. 48 These circumstances might seem to present the perfect environment for a charismatic leader to ride in and save the day by providing direction and stability. But does the economic situation lead Ukrainians to support this kind of authoritarian solution?
Chart 6.1 investigates this question by exploring the degree to which satisfaction with one’s household financial situation affects their preference for a strong leader who does not have to bother with parliament or elections. We would expect that as individuals feel that their living
46 Adam Przeworski, Democracy and the Market: Political and Economic Reforms in Eastern Europe and Latin America (New York: Cambridge University Press, 1991). 47 Taras Kuzio, “Ukraine: Leaving the Crossroads,” Central and East European Politics: From Communism to Democracy , ed. Sharon L. Wolchik and Jane Leftwich Curry (Lanham: Rowman & Littlefield, 2015), 499. 48 Ibid.
76 conditions are worse, they will be more likely to support this kind of authoritarian intervention into a system where democracy does not appear able to help them. In general , as satisfaction declines, there is a slight trend toward increasing support for this kind of strongman rule.
However, this trend is notably small and high authoritarian support persists even among those who are completely satisfied.
Chart 6.1: Strongman Rule vs. Home Financial Situation in Ukraine
Source: World Values Survey 2011
In order to further probe Przeworki’s proposition, keeping in mind that two decades have passed since Ukraine’s transition to democracy, the following will begin to investiga te whether the effect of economics on authoritarian support is caused by disappointment with democracy. If, as proposed above, defection from democracy can be traced to disappointment with its ability to improve living conditions, we might expect this effect to be stronger for those who remember
77 conditions under the previous regime. Those who are old enough to have lived through both authoritarian and democratic institutions are better poised to judge democracy in light of its alternatives. These individuals would have experienced the harsh economic downturn that occurred along with the shift to democratic rule. Obviously there are many possible intervening factors (for example: levels of repression, political performance, freedom and rule of law) so I will look for strong associations.
Table 6.1 presents correlations between satisfaction with one’s living conditions and support for a strong leader who does not have to bother with parliament or elections. I have included the same measures for Estonia to determine how Ukraine stacks up to another country with a more successful post-communist economic experience. Both surveys were taken in 2011 when each country was beginning to recover from the 2008 recession.
Table 6.1: Democracy in Comparison Age Group All Ages Under 35 Over 35 Over 50 Correlation Ukraine .045* .032 .052* .049 Between Satisfaction with Living Conditions Estonia .089** .033 .108** .113** and Rejection of Strongman Rule * significant at the .05 level ** significant at the .01 level Source: World Values Survey 2011
In both Ukraine and Estonia, there is some evidence for the idea that economic factors become more relevant for regime preference if the individual is old enough to have lived under another regime type. However, the effect of economics remains fairly low in Ukraine across age groups. Interestingly, economics seems to be a much more important consideration for
78 explaining authoritarian support in Estonia, where economic growth has been better and inflation has been kept in check.
There are some limitations to this investigation including the lack of a wide range of survey questions related to the economy in Ukraine. It remains possible that some economic factors could influence regime support. However, taken as a whole, the evidence above seems to suggest that Ukraine’s uniquely poor economic experience under democratic institutions tells us very little about the country’s high levels of authoritarian support. Financial frustration may tip the scales for a few fence-sitters as demonstrated in Chart 6.1 but support for non-democratic rule remains remarkably high no matter the economic situation.
Politics
Along with its poor economic experience, Ukraine has made only limited progress toward consolidating democracy. Numerous problems plague the country’s political institutions, and the literature (as well as the findings from Chapter 3) suggest that this should be relevant to regime legitimacy. This line of thinking assumes that Ukrainians, having experienced a deficient form of democracy, are likely to reject this type of regime in favor of an authoritarian alternative.
Several specific deficiencies could be identified as potential drivers of this kind of authoritarian support, including disrespect for human rights, failure to provide safety and security, election irregularities and widespread corruption. We expect to find perceptions of democratic failures in these areas associated with an embrace of authoritarian alternatives. This is explored in Table
6.2 below.
Perhaps the most notable finding here is that support for military rule is fairly well explained by perceptions of the political system whereas support for democracy and support for strongman rule continue to prove evasive. A perception of undemocratic governance and
79 disrespect for human rights leads some Ukrainians to support what would essentially be a military coup. A much smaller portion appears to be driven to support for strongman rule by a general perception that Ukraine is not being governed democratically.
A perception that elections are not fair is also associated with higher support for both authoritarian alternatives. This is not surprising considering the country’s history of troubled elections. In fact, after a lengthy run of fairly clean elections following the Orange Revolution, a set of 2010 elections (carried out a year before this survey) demonstrated serious flaws. 49 These flaws may have driven some Ukrainians to increasingly see military or strongman rule as a viable option, although they do not appear to have undermined democratic support.
Corruption is a persistent problem in Ukraine, which ranks 142 nd out of 175 states on
Transparency International’s Corruption Perceptions Index. 50 Wealthy business owners are presumed to profit from close ties to political leadership. 51 These real and perceived influences on politicians could be expected to undermine the legitimacy of the system. While the World
Values Survey generally lacks detailed questions dealing with corruption in Ukraine, it does ask if respondents feel that the wealthy have inordinate influence on who wins elections. This should capture at least a part of discontent with corruption, if it exists. Table 6.2 shows that a perception of inordinate influence of the wealthy in elections undermines democratic support and increases support for a military regime. However, it surprisingly has no effect on support for a strong leader who does not have to deal with parliament or elections (which has much wider support in Ukraine). Taken as a whole, the responses from this survey indicate that political
49 “Freedom in the World: Ukraine,” Freedom House, 2011, https://freedomhouse.org/report/freedom-world/2011/ukraine#.VSckipTF8vE. 50 “Corruption Perceptions Index: Ukraine,” Transparency International, 2014, accessed April, 11, 2015, http://www.transparency.org/country#UKR. 51 Ibid.
80 deficiencies are driving support for military rule to some degree but high levels of support for
strongman rule remain a mystery. I turn next to political culture in Ukraine
Table 6.2: Political Factors and Regime Support in Ukraine Support for Support for Military Rejection of Strongman Rule Rule Democracy Country Being -.040* -.128** .039 Governed Very Democratically No Respect for .012 .144** -.032 Human Rights Don’t Feel Secure in .042 -.014 -.043 Neighborhood Elections Are Not .090** .104** -.032 Fair Rich People Buy .015 .135** -.055* Elections Source: World Values Survey 2011 Culture In many studies of Ukraine, a great deal of attention is given to societal and cultural
cleavages within the country. 52 The main cleavage has its roots in a historical division of the
country between Tsarist Russia (occupying much of modern day central and eastern Ukraine)
and Austria and Hungary (western Ukraine). Ukrainians in these two regions had vastly
different historical experiences. Perhaps most relevant to the discussion of regime support is the
fact that those living in the eastern portion faced harsh autocratic rule at the hands of the Russian
Tsar while western Ukrainians were introduced to democratic institutions as early as the late 19 th century. 53 This cultural divide has been used to explain political values in Ukraine and other post-Soviet countries.
52 Paul D’Anieri, Understanding Ukrainian Politics: Power, Politics, and Institutional Design (New York: M.E. Sharpe, Inc., 2007), 105. 53 Robert Person, “Popular Support for Authoritarianism: Exploring Mass Beliefs about Democracy and Dictatorship in Contemporary Ukraine” (paper presented at the Midwest Political Science Association Annual Conference, Chicago, Illinois, April 2-5, 2009).
81 It has often been argued that Russian culture is characterized by several authoritarian
qualities including fatalism, intolerance, orthodox religion and the importance of the group over
the individual. 54 Most importantly, Russian culture has been noted for its preference for strong leadership linked to the traditional peasant culture and the necessity of strong central rule to govern the country’s vast territory. 55 Conversely, Ukrainian culture, particularly in the west, is more often linked to liberal European values. Therefore, we have good reason to suspect that the significant preference for strongman rule in Ukraine is a remnant of Russian culture and we should find it at its strongest among those who identify as ethnically Russian, speak Russian or live in eastern Ukraine.
The charts below demonstrate that this is not the case. In fact, Russians and eastern
Ukrainians are very slightly less supportive of strongman rule across the board. The trend of support for this authoritarian option seems to cross regional and historical boundaries, encompassing the entire country. This might lead one to reject the idea that culture is influencing authoritarian support, but a closer look at national-level trends reveals that that would be unwise.
54 William M. Reisinger, Arthur H. Miller, Vicki L. Hesli and Kristen Hill Maher, “Political Values in Russia, Ukraine and Lithuania: Sources and Implications for Democracy,” British Journal of Political Science , 24(2) (1994) 183-223. 55 Ibid.
82 Chart 6.2: Support for Strongman Rule vs. Ethnicity
Source: World Values Survey 2011
Chart 6.3: Support for Strongman Rule vs. Language
Source: World Values Survey 2011
83 Chart 6.4: Support for Strongman Rule vs. Ethnicity
Source: World Values Survey 2011
Using Christian Welzel and Ronald Inglehart’s “emancipative values” index, we find that
Ukraine is characterized by a relatively authoritarian culture. Emancipative values are scaled from 0-1 and countries or individuals with high levels are characterized by giving priority to
“equality over patriarchy, tolerance over conformity, autonomy over authority, and expression over security.” 56 While countries like Estonia and Poland that reject authoritarian alternatives receive average ratings of .48 and .46 respectively, Ukraine’s average was only .40, the same as
Russi a’s. This suggests that we cannot reject the idea that Russian culture is somewhat more prone to authoritarianism. It also seems to indicate that Ukrainian culture may be closer to
Russia than to Europe with respect to authoritarian and democratic values .
56 Christian Welzel and Ronald F. Inglehart, “Political Culture, Mass Beliefs, and Value Change,” in Democratization , ed. Christian W. Haerpfer, Patrick Bernhagen, Ronald F. Inglehart, and Christian Welzel (New York: Oxford University Press, 2009), 129.
84 Chart 6.5 shows the effect of emancipative values on support for strongman rule, demonstrating that cultural differences have a stronger impact in Ukraine than in Estonia or
Poland. In Ukraine, a shift from the highest to the lowest level of emancipativ e values corresponds with a nearly 60-point drop in the percentage of Ukrainians who support strongman rule. This measure of culture appears to be a factor in all three countries included but it remains particularly important for Ukraine, where authoritar ian support is much higher to begin with.
Chart 6.5: Support for Strongman Rule vs. Emancipative Values
Source: World Values Survey 2011
Drilling down even further on the nature of regime support in Ukraine sheds light on the huge percentage of the popu lation that approves of strongman rule. The data from the 2011
World Values Survey in Ukraine reveals a striking trend: Ukrainians do not necessarily see democracy and rule by a strong leader ignoring elections and parliament as mutually exclusive.
85 Table 6.3 shows the correlations between support for democracy and support for strongman rule in several Central and Eastern European states. We should expect to see that individuals who support democracy as a regime type would widely reject strongman rule. Instead we find that this is only the case in Estonia and Poland, the two countries that more willingly reject this authoritarian alternative. In Ukraine there is no association between the two. In fact, 59% of
Ukrainians approve of both democracy and strongman rule as viable regime options for the country. In Romania there is a significant correlation linking both types of regime support.
Table 6.3: Support for Democracy vs. Support for Strongman Rule Ukraine Romania Poland Estonia Correlation -.038 .082* -.140** -.187** Between Support for Democracy and Support for Strongman Rule Source: World Values Survey 2011
Support for strongman rule is also closely tied with support for the government in power.
This seems to indicate that many Ukrainians are comfortable with their political leaders taking full advantage of and consolidating their position of power, as many have done. The Chart below demonstrates that this is true not only in 2011 as Viktor Yanukovych returned to the political scene and began to recentralize presidential power, but also in 2006, just two years after the mass uprising during the Orange Revolution. At this point, Viktor Yushchenko was still in power after being elected in a fair electoral transfer of power. This indicates that for some, a vote for Yushchenko was not a vote for deepening democracy per se but rather a vote for a
“better” strongman ruler.
These trends in mass political values are somewhat reflected in Ukraine’s political institutions. Political parties in Ukraine tend to be used as personal vehicles by their leaders
86 rather than adhering to coherent platforms or ideologies. 57 With popular demands matching political institutions, it seems unlikely that we will see significant progress toward democracy and institutiona lization of checks and balances against a traditionally strong Ukrainian president.
This is certainly a surprising finding considering the democratic excitement that surrounded the
Orange Revolution. We will have to wait to see what changes are brought b y the significant recent developments of mass protests and civil war.
Chart 6.6: Strongman Rule vs. Confidence in Government
Source: World Values Survey 2011
Conclusion The Orange Revolution in Ukraine brought a great deal of optimism and it seemed to demonstrate a thriving democratic culture in a weakly democratic holdout country. However, survey data casts doubt on this interpretation showing that high levels of support for
57 Freedom In The World: Ukraine 2011.
87 authoritarianism persist despite mass mobilization in favor of democracy. In particular, large majorities are in favor of a regime where a strong leader ignores parliament and is not subject to elections. This chapter sought to explain this puzzle building on the framework of the preceding chapters.
Economic and political variables offer little help in explaining high levels of support for strongman rule in Ukraine. Survey data showed that satisfaction with one’s living conditions is only weakly associated with rejection of this autocratic alternative. Various measures of the success and failings of democracy itself offered very little help. These political variables did however prove useful for explaining varying levels of support for military rule.
Culture stands out as a central factor in explaining regime support in Ukraine. Welzel and Inglehart’s Emancipative values show that the country is characterized by a political culture closer to Russia than Estonia or Poland. Interestingly, unlike many aspects of Ukrainian politics, authoritarian support cannot be divided cleanly into ethnic, linguistic or regional categories.
Support for strongman rule seems to have penetrated each corner of the country. What’s more, this authoritarian support is not disconnected and may even be an important part of support for democracy in Ukraine. There is compelling evidence here that many Ukrainians desire both democracy and a strong leader who does not have to deal with it. Whether these popular preferences can be reconciled is uncertain and may help explain some of the instability and weak institutionalization of the country’s democracy. What will be brought by newer developments in the country is uncertain, but the cultural source of authoritarian support in Ukraine would seem to present a difficult obstacle toward further democratization.
88 Chapter 6: Authoritarian Support in Guatemala and Nicaragua
Democracy is struggling to take root in two Central American countries, Nicaragua and
Guatemala. Both received a score of 4.0 from Freedom House’s annual Freedom in the World
measure in 2011, recognizing them as only partly free. Widespread corruption wreaks havoc on
democratic institutions and diminishes citizen trust in both countries. Transparency International
ranked Guatemala 120 th and Nicaragua 134 th of 180 countries in its 2011 report on Corruption
Perceptions. Cases of graft continue to reach to the highest levels of government and journalists are increasingly fearful of repercussions for taking on controversial stories. 58 In both countries, civil society faces harassment, rights guarantees are not always respected for indigenous communities and violence against women is prevalent. 59
Weak institutions are unable (or officials are unwilling) to address deep socioeconomic issues that have persisted for decades. First among these are the highest levels of poverty in the region. The percentage below the national poverty line was 53.7% in Guatemala and 42.5% in
Nicaragua in 2011 and 2009 respectively. 60 Most estimates put one of these two countries as the most impoverished in Central America. This is compounded by a high concentration of wealth in the hands of the economic elite. Both countries remain underdeveloped and dependent on the international economy, making the poor even more vulnerable. Beyond politics and economic issues, both Nicaragua and Guatemala have very young populations, roughly 72% literacy rates,
58 “Freedom in the World: Nicaragua,” Freedom House, 2011, https://freedomhouse.org/report/freedom-world/2011/nicaragua#.VQpRLWTF8vE; “Freedom in the World: Guatemala,” Freedom House, 2011, https://freedomhouse.org/report/freedom- world/2011/guatemala#.VQfB32TF8vE. 59 Ibid. 60 World Development Indicators, The World Bank, 2012, Washington, D.C: The World Bank, http://data.worldbank.org/indicator/SI.POV.NAHC/countries?page=1.
89 a similar proportion of the population living in cities (57.5% in Nicaragua compared to 49.8% in
Guatemala) and the average citizen spends 11 years in school. 61
Despite many similarities however, these two countries differ significantly in the levels of their populations’ preference for authoritarian rule. Based on survey data described below,
Guatemalan citizens appear much more willing to accept some form of authoritarian rule compared with Nicaraguans.
Using the simplest measure, Latinobarómetro data from a survey carried out in 2011 shows that 24.7% of Guatemalans would support an authoritarian government under some circumstances compared with 17.8% of Nicaraguans. The same survey found that 36.7% in
Guatemala disagreed that “democracy may have problems but it is still the best form of government,” compared with only 16.2% in Nicaragua. Adding depth to that finding is a separate survey carried out by the Latin American Public Opinion Project (LAPOP) in 2010 that asks the same question but with a 7-point scale of agreement rather than simple categorical answers. According to this survey, in Guatemala, a tiny 17.93% “very much agree” that democracy is better than any other form of government compared to 44.38% in Nicaragua.
These disparities certainly point to a stronger willingness to embrace regime alternatives in
Guatemala. This finding persists (and is even amplified) when observing attitudes toward a specific regime alternative; military rule. Support for the prospect of a “military government in replacement of a democratic one if things got very bad” reaches 56.6% in Guatemala compared with only 19.9% in Nicaragua. What explains these differences in authoritarian support? Many of the factors identified in previous chapters offer little to help explain this puzzle. The countries share similar political systems, economic conditions and legacies of repression.
61 The World Factbook, Central Intelligence Agency, 2014, https://www.cia.gov/library/publications/the-world-factbook/geos/nu.html.
90 The following analysis suggests three main differences that are driving disparities in regime legitimacy. First, the experience of social revolution in Nicaragua (not present in
Guatemala) seems to have contributed to comparatively high levels of democratic support and slightly lower levels of authoritarian support. Second, the presence of large indigenous population in Guatemala may generate resistance to their full inclusion by the traditionally powerful racial groups. Thirdly, high levels of crime and violence in Guatemala as compared to
Nicaragua appear to generate desire for authoritarian solutions.
Political and Economic Legacies
A review of the history of the two countries under observation will help put future developments in context and recognize underlying currents in politics. I argue that although
Nicaragua and Guatemala have diverging histories, they share a number of important historical legacies that continue to shape politics today. Among these are the development of a plantation society, a history of authoritarianism, strong foreign influence, socioeconomic crisis, peasant and labor resistance, and harsh repression.
During the colonial and post-independence period, both countries developed as plantation societies eventually producing primarily coffee. This brought with it a large peasant class that suffered from encroachment and exploitation by a land-hungry elite. Economic conditions for this segment of society continue to be dismal and the agricultural sector in both countries is still important. Additionally, the integration of the two countries into the international economy made the peasant classes particularly vulnerable to global economic shifts and crises.
The 20 th century in both countries was characterized by harsh authoritarian rule and strong foreign (mostly US) influence. From roughly 1871-1944 Guatemala was ruled by a series
91 of military dictators aligned with the coffee oligarchy and foreign investors. 62 The October
Revolution in 1944 heralded in 10 years of democracy that brought social programs and land reform to the country for the first time in its post-conquest history. However, this brief interlude was ended by a CIA organized ouster of the sitting president that led to years of harsh military rule and civil war. The end of this democratic period in Guatemala also meant the dismantling of the reforms that had been gained by it. Foreign influence was even more overt in Nicaragua where the US physically occupied the country from to 1912 to the 1930s. The exit of the US army brought about the decades-long rule of the corrupt and repressive Somoza family.
Elite rule and growing international demand for coffee led to a worsening situation for the rural poor in both countries. Economic conditions declined for peasants and the working- class while rich landowners grew wealthier. Several attempts at organized resistance occurred in both countries including the October Revolution, mentioned previously, and a period of indigenous organization and insurgency during the 70s and 80s in Guatemala. In Nicaragua, the rebellion of guerilla leader Augusto Nicolás Sandino brought about the armed conflict that resulted in the end of foreign occupation of the country. The next rebellion, some 30 years later, would take Sandino’s name and build upon his strategy with considerable success. This
Sandinista rebellion will be discussed further on. What unites all of these examples, as well as smaller-scale, more peaceful peasant organization, is that they were often met with harsh, violent repression from the ruling regimes.
Despite many shared historical legacies, including harsh repression that should have generated distaste for authoritarian rule in both countries, differences in authoritarian support
62 Harry E. Vanden and Gary Prevost, Politics of Latin America: The Power Game (New York: Oxford University Press, 2006), 267.
92 persist. Clearly there are more factors at play including, perhaps, the point of departure for
Guatemala and Nicaragua’s shared history; social revolution in Nicaragua.
Social Revolution vs. Protracted Civil War
The histories of Nicaragua and Guatemala experienced a significant divergence in 1979 when revolutionaries in Nicaragua, grouped under the FSLN (the Sandinistas), forced Anastasio
Somoza DeBayle into exile and took control of the government. Nicaragua had experienced a social revolution, uniting laborers, businesses, students, opposition parties and the Catholic
Church against the ruling family. Guatemala, never managing to unite such a strong opposition, saw only an increasingly brutal civil war as the insurgency there was met with the growth of repressive institutions. Still, civil war in Guatemala does not make the country particularly exceptional. In fact, following the revolution in 1979, Nicaragua faced a similarly violent civil war, as did El Salvador in the 80s and 90s. The factor that differentiates the two countries in this case is the unique experience of revolution in Nicaragua that never occurred in Guatemala.
Perhaps this revolutionary experience left a lasting mark on the country that may account for its lower levels of authoritarian support. I will explore two possible manners through which the popular revolution may have affected authoritarian support in Nicaragua: the promotion of democratic norms and beliefs that undermine support for illiberal regimes and a Sandinista education program that demonized of the previous regime.
The revolution in Nicaragua brought efforts to improve health, widen education, fairly distribute land, and empower local mobilization in a new form of grassroots democracy. While subsequent regimes shifted away from these reforms, it was nothing compared to the complete reversal experienced in Guatemala after the brief democratic interlude brought on by the October
Revolution. Guatemala’s two reformist leaders during this period also made efforts to widen
93 education, promote public health and encourage labor and peasant organization. The early 1950s
saw the redistribution of farmland to 100,000 peasants. 63 However, the US-backed coup in 1954
brought to power Colonel Castillo Armas, who undertook the complete dismantling of these
reforms. As one account puts it: “[w]ith ferocious anti-Communist propaganda, the
counterrevolution dismantled the labor and peasant movements, killed and jailed thousands,
repressed political parties, revoked the Agrarian Reform Law, and returned confiscated lands to
their former owners.” 64 Additionally, this brief democratic period occurred more than half a
century ago in what may now be a distant past.
While the Sandinista regime was not the perfect model for democracy, it must have at
least felt democratic compared to years of repressive rule. And that democratic experience led to
some important lasting changes, possibly demonstrating to citizens that democracy can make a
real difference. Additionally, one should not underestimate the ideological power of the mass
mobilization that brought about revolution. The combination of the experience of revolution and
the realization that democracy may be able to solve the country’s problems could certainly be
expected to raise support for a liberal regime, thereby diminishing authoritarian support. Adding
to a mixed literature on the topic, John A. Booth and Patricia Bayer Richard found, using a
comparative lens, that the experience of revolution in Nicaragua left a lasting impact on levels of
leftist identification, involvement in educational groups and the way that communal groups
mobilize people into political campaigns. 65 It did not, they argue, have an impressive impact on electoral participation or, somewhat surprisingly, communal organization on its own. They also
63 John A. Booth, Christine J. Wade and Thomas W. Walker, Understanding Central America: Global Forces, Rebellion and Change (Colorado: Westview Press, 2015), 175. 64 Ibid. 65 John A. Booth and Patricia Bayer Richard, “Revolution’s Legacy: Residual Effects on Nicaraguan Participation and Attitudes in a Comparative Context,” Latin American Politics and Society , 48(2) (2006).
94 note that the revolution must have led to greater support for democratic values in society,
although without much evidence to support this claim.
A final reason that we may expect lower authoritarian support in Nicaragua has to do
with the educational reforms pursued by the Sandinista regime in the wake of the revolution.
Education reform was a central aspect of the Sandinista project and efforts to improve literacy
began right away. This education program was, at least in part, ideologically driven and propped
up the FSLN regime at the expense of prior rulers. As one observer puts it, “Sandinista rhetoric
and heroes were promoted through textbooks, and reaction was often particularly harsh about the militarism and violence depicted in teaching materials.” 66 The opportunity of the Sandinista’s to
“rewrite history”—a history that would certainly portray the Somoza regime in a very harsh
light—could further reduce the number of individuals that would want to return to that kind of
rule.
In sum, the experience of revolution may be expected to have mobilized a democratic
citizenry and demonstrated the utility of democratic institutions. It also may have played a part in
inducing fear of returning to the past regime through a mass education campaign. Chart 5.1
investigates these first two claims further by comparing levels of support for democracy over a
15-year period across Central America. Staying consistent with Booth and Richard’s
comparative framework, in order to ensure that higher levels of support in Nicaragua can be
traced to its unique experience of social revolution, I have included two other Central American
countries, El Salvador and Honduras, in this analysis. El Salvador’s political experience was
similar to Guatemala’s with a lengthy civil war during the late 1900s. Honduras escaped this
level of conflict but its population still lived under authoritarian institutions during this period.
66 Merilee S. Grindle, Despite The Odds: The Contentious Politics of Education Reform (New Jersey: Princeton University Press, 2004), 77.
95 Nicaragua was the only country to experience a successful revolution so if the predicti ons outlined above prove correct, we should see the highest levels of democratic support there.
Chart 5.1: Democratic Support in Central America
Source: Latinobarómetro 2011
In general, Nicaragua does display higher average support for democracy across the period observed. However, the difference is somewhat limited with a 57.4% average in
Nicaragua over the 15-year period compared to 53.6%, 50.4%, and 40.3% averages in El
Sa lvador, Honduras and Guatemala respectively. Additionally, Chart 5.1 shows some variation over time with Nicaragua at times falling below the other Central American countries in terms of democratic support. Guatemala continues to be an outlier with far a nd away the lowest levels of democratic support in the region.
Now the question becomes whether the slightly elevated levels of support for democracy in Nicaragua reflect a rejection of authoritarian rule or simply lower levels of indifference to
96 regime type. Chart 5.2 reflects the percentage of respondents in each country who, in response to the same question used for democratic support, chose instead that they would support an authoritarian regime under some circumstances. Other possible response options were indifference to regime type, “don’t know” or “didn’t answer.” As demonstrated in previous chapters, low democratic support does not always translate into high authoritarian support.
Consistent with levels of support for democracy, Chart 5.2 reveals slightly lower support for authoritarianism in Nicaragua compared to other countries in the region. Nicaragua’s levels of authoritarian support across this period average 12% compared with 13.6% in El Salvador,
14% in Honduras and 22.2% in Guatemala. However, this difference is even less pronounced than the differences in democratic support. Again, there is significant variance over time and
Nicaragua is not always the country with the lowest authoritarian support. Interestingly, authoritarian support appears to be much more volatile than democratic support suggesting that it may, in some cases, represent a response to a difficult situation rather than a more entrenched idea, as democratic support appears to be. Still, there are significant differences between these countries that form trends rather than one-off events. Over this period, support for authoritarianism in Guatemala remains the highest in the region.
Returning to a comparative analysis of Guatemala and Nicaragua, there does appear to be higher general support for democracy in the latter, likely a result of the experience of revolution.
This is also true of authoritarian support, the subject of this investigation, but the effect is slight
(especially with respect to other regional countries). The fact that Guatemala remains an outlier suggests that the experience of revolution is only a small part of the explanation for the differences in authoritarian support between the two countries. Another variable that may help explain this puzzle is the differing effect of inequality in each country.
97 Chart 5.2: Authoritarian Support in Central America
Source: Latinobarómetro 2011
Inequality
Chapter 2 identified a trend towards countries with greater inequality having higher levels of authori tarian support. Some of the data se emed to suggest that this was related to Carles
Boix’s assertion that unequal distribution of wealth in a country will drive some to violate the rules of the democratic game. In the context of this investigation, that w ould mean that established elite would prefer authoritarian rule as a means to preserve their wealth. We may be able to gain greater insight into this proposition, and whether it relates to the two countries under investigation here, by looking at the way that different social classes feel about authoritarian rule.
Guatemala is slightly more unequal than Nicaragua; the former scored 52.4 on the GINI index compared with the latter’s 45.7. 67 Still, these countries both demonstrate fairly high levels
67 The World Bank. “ World Developm ent Indicators.” The World Bank , 2012. Accessed on April 17, 2015. http://data.worldbank.org/indicator/SI.POV.NAHC/countries?page=1 .
98 of ineq uality. However, Chart 5.3 shows that the effect of social class on authoritarian support is slightly different between the two countries . Class is measured by a self-reporting on a scale of
1-10, with 1 being the poorest and 10 being the richest.
Chart 5.3 : Authoritarian Support vs. Economic Class
In neither case does class appear to be the primary driver of support for authoritarianism.
In Guatemala, levels of support for authoritarianism are fairly consistent across the different points on the sca le of wealth. The same is roughly true in Nicaragua, but here we can see a slight trend toward higher authoritarian support among the wealthier citizens. Among the richest tenth,
26.7% support the idea of an authoritarian regime, while the same is true o f only 10.7% of the poorest Nicaraguans. This seems to offer some evidence for the idea that, in some cases, inequality can drive those who would be threatened by evening the playing field to support an authoritarian regime option in order to secure their benefits. However, it does little to explain
99 why authoritarian support is so much higher in Guatemala. To continue to investigate this puzzle,
we will next turn to each countries ethnic composition.
Indigenous Population
Another important difference between Guatemala and Nicaragua is the presence of a
significantly large indigenous community in the former compared to a relatively homogenous,
mestizo population in the latter, with only a small, geographically isolated native population. In
Guatemala, indigenous people make up roughly half of the population. In Nicaragua the
proportion is closer to 5%. To probe the extent to which these differing racial makeups
contribute to regime legitimacy, tables 5.1 and 5.2 demonstrate the relationship between
ethnicity and support for authoritarianism in both countries.
Table 5.1 shows that respondents with Indigenous ethnic backgrounds were more likely than all other major ethnic groups to reject authoritarianism. Only 19% of native Guatemalans were willing to accept some form of authoritarianism compared with 33.7% of Mestizos and
23% of Whites. In contrast to Guatemala, the pattern goes the other way in Nicaragua where
Indigenous people tended to be more accepting of authoritarian rule. Here, 21.7% of natives reported conditional preference for authoritarianism compared to 13.2% of Mestizos, 12.5% of
Whites and 3.6% of Blacks. Although these differences are small in both cases, they do appear to demonstrate important patterns in both cases. 68 Support for authoritarianism in Guatemala
68 Why are indigenous people more willing to accept authoritarianism in Nicaragua than in Guatemala? One explanation for this might lie in the fact that indigenous Guatemalans received the brunt of military repression during the harshest crackdown of the 70s and 80s. The specific racial targeting of government violence in this case means that large portions of the population might not see authoritarian rule as a threat based on their own past experience with it. This could help explain the fact that a greatly improved record of repression under democracy has not translated into rejection of authoritarianism in the same way that it has in other countries outlined in Chapter 4. Another explanation could be the differing participation of indigenous people in revolutionary movements during the late 20 th century. Indigenous groups in Guatemala tended
100 remains at concerning levels no matter your race. These levels become much more concerning for non-indigenous Guatemalans. Perhaps the question is not why indigenous groups are less supportive of authoritarianism, but rather why non-indigenous groups are more supportive of alternatives to democracy. The explanation here may lie in patterns of racism that underlie
Guatemalan politics.
Over, three-quarters of Guatemalans view their society as discriminatory toward the indigenous population. 69 Unlike in Nicaragua, indigenous people, if united, could form a powerful political force if granted full participation rights. It may not be a leap to think that a discriminatory environment combined with a potentially powerful yet suppressed population could lead some Guatemalans in the “in-group” (mestizo and white) to harbor reservations about the full inclusion that democracy entails. A lack of comparable survey data on racist attitudes in
Guatemala limits our ability to test this proposition but there may be some anecdotal evidence pointing in this direction.
As part of the ongoing Peace Accords following the end of the 30-year civil war,
Guatemalans voted in a 1999 referendum including important provisions on indigenous recognition, rights and language usage. In the context of incredibly low turnout, the referendum was rejected in a fairly close vote. Close inspection of the election results reveals that by far the highest concentration of “No” votes was in urban and mestizo areas, while the “Yes” vote was more common in areas with higher concentrations of indigenous peoples. 70 Among competing
to participate in revolutionary activity while many natives in Nicaragua were involved with the counterrevolutionary “Contra” forces. 69 Latinobarómetro Corporation, 2011, "Latinobarometro / Latinobarometer Data Files", http://hdl.handle.net/1902.29/10528 Latinobarometro Corporation [Distributor] V1 [Version]. 70 Dinorah Azpuru, “The Consulta Popular: A Vote Divided by Geography,” The Popular Referendum And The Future of the Peace Process in Guatemala , Wilson Center Working Paper Number 214 (1999), 13.
101 Table 5.1: Guatemala Asian Black Indigenous Mestizo Mulato White Other Democracy 34.6% 21% 37.5% 38.4% 33.5% 36.1% 48.7% always preferred Sometimes - 33.8% 19% 33.7% 33.5% 23% - Authoritarian rule prefered Doesn’t Matter 65.4% 22.6% 28.9% 24.8% 33% 32.8% 34.7%
Didn’t Answer - - 1.4% - - 0.7% -
Don’t Know - 22.6% 13.2% 3.1 - 7.4% 16.6% (N) (3) (9) (450) (154) (6) (291) (6) Source: Latinobarómetro
Table 5.2: Nicaragua Asian Black Indigenous Mestizo Mulato White Democracy 37.6% 53.5% 51.9% 52.7% 70.2% 57.9% always preferred Sometimes 24.3% 3.9% 21.7% 13.2% 17.6% 12.5% Authoritarian rule prefered Doesn’t Matter 24.9% 31.5% 8.5% 18.9% 6.1% 22.1%
Didn’t Answer - - 1.3% 1.0% 6.1% -
Don’t Know 13.2% 11.1% 16.6% 14.2% - 7.5% (N) (8) (27) (81) (670) (17) (55) Source: Latinobarómetro 2011
explanations for this phenomenon is a fear of change and even outright racism. In a 1999 news
article analyzing the results, Juan Hernandez Pico noted that in the debate in the media running
up to the vote, “a large sector of the ladino population feared that the constitutional reforms would shift the balance of power in Guatemala, starting a dangerous trend by recognizing the majority status of the Mayan population and their cultural values, spirituality, common law and
102 languages.”71 He points to a “hidden racism” that exists not in the form of overt discrimination but rather a convenient pattern of denial of full status and freedoms to indigenous people.
While it is certainly possible that this kind of racism may contribute to higher levels of authoritarian support in Guatemala—and the experience of the referendum appears to back this thinking up—the lack of hard data prevents any solid conclusions. In any case, the ethnic divisions in support for authoritarianism do appear relevant but still do not seem to capture the full story. The final variable investigated below is the pervasive violence in Guatemala that does not reach comparable levels in Nicaragua.
Violence and Crime
Violence and organized crime have become pervasive in Guatemala. The murder rate in
2010 was 42 per 100,000 citizens compared with 14 in Nicaragua. Weak institutions and the legacy of the civil war have contributed to a sense of impunity in Guatemala as only 2% of crimes reported to the Public Ministry ever see their day in court. 72 As Guatemalans lose faith in the ability of their government to protect their security, they could well be expected to embrace authoritarian solutions that would at least bring calm. The election of Conservative President
Otto Pérez Molina can be seen as a step in this direction. 73 Molina ran against a leftist incumbent in 2012 promising a return to mano dura , or iron fist, politics in order to get the escalating violence under control. Additionally, high levels of military support in Guatemala may reflect a willingness to accept a military solution to these systemic problems.
71 Juan Hernandez Pico, “Why Was The Referendum Defeated?” Envio Number 216 (1999). 72 Anita Isaacs, “Countries at the Crossroads: Guatemala,” Freedom House (2012), https://freedomhouse.org/report/countries-crossroads/2012/guatemala#.VRCzQ2TF8vE. 73 Gregory Weeks, Understanding Latin American Politics (New York: Pearson, 2015), 120.
103 There is indeed some scholarly evidence that the kind of criminal violence that exists in
Guatemala—but is much less present in Nicaragua—negatively impacts regime legitimacy. 74
For example, using LAPOP survey data, Miguel Carreras finds that both crime victimization and perceptions of criminal violence lead to lower system support. He notes, as suggested above, that this disenchantment may promote support for authoritarian solutions. Using
Latinobarómetro data, Chart 5.4 examines this idea in Nicaragua and Guatemala. It demonstrates that, in Guatemala, the more concerned an individual is about being a victim of a violent crime, the more likely they are to support the idea of an authoritarian regime. Only
16.7% of those who never feel threatened would support authoritarianism under some circumstances compared with 26.6% of those who feel threatened all or almost all of the time.
Interestingly, Nicaragua displays the reverse pattern with individuals that feel more secure actually being slightly more likely to support authoritarianism. These findings offer some support to the idea that the pervasive nature of violence in Guatemala makes it an important factor in the determination of regime support. The same cannot be said for Nicaragua where violence has not penetrated daily life to the same degree.
Related to the widespread violence in Guatemala is the issue of increasing penetration of the state’s institutions by drug traffickers and organized criminal syndicates. Large-scale government scandals in Guatemala often have undertones of this kind of penetration, a situation not so visible in Nicaragua. For example, the February 2007 killing of three Salvadoran representatives to the Central American Parliament was eventually traced to a Guatemalan
74 Miguel Carreras, “The Impact of Criminal Violence on Regime Legitimacy in Latin America,” Latin American Research Review , 48(3) (2013); Orlando J. Pérez, “Democratic Legitimacy and Public Insecurity: Crime and Democracy in El Salvador and Guatemala,” Political Science Quarterly , Volume 118, Number 4 (2004).
104 Chart 5.4: Crime and Authoritarian Support
Source: Latinobarómetro 2011
politician and four Guatemalan policemen including the head of the organized -crime unit. 75
Additionally, in 2005, the head of the anti -drug agency in Guatemala was arrested by the US on charges of drug trafficking. 76 Notably, this event corresponds with a massive jump in authoritarian support in the country between 2005 and 2006 that can be ob served in Chart 5.2.
Although Nicaragua is no stranger to high -profile corruption cases and shaming of politicians, the absence of the pronounced criminal organizations found in Guatemala makes allegations perhaps less d amning in the former. Chart 5.5 may shed some light on the different ways that citizens perceive corruption in both countries.
75 Booth, Wade and Walker, 189. 76 Mark Sherman, “Guatemalan Anti -Drug Official Indicted,” The Washington Post , November 16, 2005, accessed March 23, 2015, http://www.washingtonpost.com/wp- dyn/content/article/2005/11/16/AR2005111601349.html.
105 Chart 5.5 : Corruption and Authoritarian Support
Source: Latinobarómetro 2011 The findings are quite clear: those who have heard of an act of corruption in Guatema la are almost three times more likely to support authoritarianism than those who have not. In
Nicaragua, the reverse is true but the effect is much smaller. I would argue that this is due to the differing nature of corruption in the two countries. Corru ption in Guatemala tends to demonstrate deeply concerning links between the government and criminal groups , taking the form of organized crime . The fact that criminal violence is a central concern in Guatemala makes this situation even worse. How can the current regime be expected to solve the most pressing issues in the country if they are colluding with the enemy? In Nicaragua, corruption may demonstrate government greed, incompetence, trickery and theft, but it does not present an immediate threat to citizens in the same way that it does in Guatemala. The persistence of
106 violence and the apparent government complicity in it in Guatemala may offer the best
explanation yet for its differing levels of authoritarian support.
Conclusions
Despite many similarities, Guatemala and Nicaragua display consistently different levels of authoritarian support. Guatemalans appear much more open to some form of authoritarian rule than their Nicaraguan counterparts. Explanations pursued in previous chapters offer little help in explaining this variance as both countries share similar political regimes and socioeconomic conditions. Both countries have also seen a decrease in repression under nominally democratic institutions. In explaining regime support in these two countries, we are forced to look outside of the mold laid out in the literature.
The historical experience of social revolution in Nicaragua seems to have elevated democratic support in that country thereby slightly diminishing authoritarian support.
Nicaragua’s homogenous mestizo population may also contribute to its comparatively low levels of authoritarian support. In Guatemala, where native populations make up around half of the population, non-indigenous racial groups appear more supportive of authoritarian alternatives.
This may be due in part to xenophobia and even racist fears of allowing a historically marginalized group into power by fully embracing democracy. This analysis found anecdotal evidence to support this proposition but the limitations of the survey data available prevent a closer analysis. Class differences do not appear to drive a similar fear of capture of the system by the masses in Guatemala, although wealthier individuals in Nicaragua tend to demonstrate slightly higher authoritarian support. Finally, Guatemala’s higher levels of violence and patterns of government collusion with criminals appear to play a part in fostering support for authoritarian solutions where democracy has failed.
107 In general, it is clear that support for authoritarianism has multifaceted explanations.
There is unlikely to be any one-size fits all explanation and there even appears to be significant variation in explanations across different countries. This suggests that debates between global or regional analysis of regime support may be somewhat misplaced. A smaller-scale investigation of individual countries appear to be the best way to proceed when looking to explain trends in support for authoritarianism. This investigation seems to have revealed that finding issues that are of particular importance to or cause significant contention among the population (like public security and indigenous rights in Guatemala) may be the best way to root out the causes of regime support.
108 Chapter 7: Conclusions
This thesis sought to explain varying levels of authoritarian support across many new (or
relatively new) democracies. I uncovered that levels of authoritarianism vary widely across
Latin America, Africa and Central and Eastern Europe. Looking closer at support for specific
authoritarian alternatives reveals even starker differences.
Support for authoritarianism is the comparatively understudied flip side of democratic
legitimacy. For this investigation, I took the fairly wide-reaching literature on democratic
support and attempted to apply it to the more peculiar condition of authoritarian support. This was aimed at confirming previous research, attempting to add my own insights, and seeing to what degree these two measures of popular support are actually related.
I have found significant variation in the triggers of authoritarian support across different regions. Therefore, while the structure of this thesis was divided by particular issues (political, economic and cultural), this conclusion will be broken down by region. In this chapter, I will describe my main findings for Latin America, Africa and Central and Eastern Europe. I will then illuminate some overarching trends. Stepping back to look at this research as a whole I find two important insights that could help guide future studies of regime support. First, authoritarian support should be viewed as distinct from democratic support. High support for democracy can sometimes be deceiving and low support for democracy can indicate indifference rather than acceptance of authoritarian institutions. Second, authoritarian support seems to be best analyzed at a smaller scale, looking at the unique contexts of individual countries.
Latin America
In Latin America, politics seem to take priority over economics in determining support
for authoritarianism. Still, some aspects of modernization appear to be relevant. Urbanization
109 and increasing education are linked to decreasing support for authoritarian alternatives. These processes may introduce democratic values or perhaps help prevent charismatic autocrats from taking advantage of a more widely spread, less educated population. Inequality and poverty may contribute to support for autocracy as well but, particularly with respect to inequality, the evidence is contradictory. Although countries identified as being more unequal by the GINI index tend to have higher levels of popular authoritarian support, this is not mediated through individuals’ perceptions of wealth distribution. It remains possible that some intervening factor is behind this. Future studies could investigate related issues like land distribution or elite-level corruption.
While many deeper economic conditions seemed at least partially relevant to authoritarian support in Latin America, shorter-term evaluations of economic conditions were not as important. This seems to indicate that economic downturn is not likely to result in large portions of the population of Latin American countries jumping ship and shifting their loyalty to authoritarian alternatives. This study stopped short of investigating whether deep, prolonged economic crisis might have that result.
Various measures of the political system in Latin America tended to be better predictors of regime support than shorter-term economic factors. General evaluations of democracy’s progress and levels of satisfaction with how democracy is working were significantly connected to support for authoritarianism. More specifically, high levels of corruption are consistently connected with support for autocracy. These correlations were generally stronger than those found for economic measures but they were still notably small and left the question of authoritarian support only partially answered.
110 An important finding in Latin America was that a history of repression, especially when the new democratic regime is seen as an improvement over the past, is associated with greater rejection of authoritarianism. Many in the region seem to have learned the historical lesson and do not desire that the country end up in a similar situation again. This returns us to the notion, highlighted more than a decade ago by Richard Rose and his colleagues, that authoritarian alternatives are evaluated with respect to familiar forms of government. This seems to imply that citizens have concrete associations with different forms of authoritarian rule rather than some abstract idealized picture.
Although it is merely speculation, some of the other major findings for the region could be interpreted through this lens. For example, if corruption levels are seen as having increased under democratic rule (a system where there are more “palms to grease”), then authoritarianism may be seen as comparatively clean form of rule, even if this might not really be the case. As another example, the effect of inequality on authoritarian support could be related to a perception that land reform is difficult under a democratic regime. This framework opens up many interesting lines of study that could continue to uncover the sources of authoritarian support in this region.
Africa
Demonstrating the very real differences between these three regions, this thesis has all but failed to solve the puzzle of authoritarian support in Africa. Neither economic nor political factors explained more than a miniscule amount of variance in support for autocracy. Culture was equally inconclusive, showing in some cases the opposite of the expected effect on regime support.
111 Still, the very unpredictability of authoritarian legitimacy in this region is fascinating.
Why is Africa so resistant to the same factors that are turning populations toward autocrats and military rule in other regions? Perhaps applying a similar framework to the one enumerated above for Latin America could bear interesting results in Africa. This might involve looking closer at the nature of past authoritarian regimes in the region to see what conditions might dissuade populations from considering a return to that system of rule.
Another framework that has been used in the past for tackling the puzzle of authoritarian support in Africa involves looking at each country on a case-by-case basis. Michael Bratton,
Robert Mattes and E. Gyimah-Boadi exemplify this method in their analysis of regime support in
Public Opinion, Democracy, and Market Reform in Africa . As an example, they attribute high support for one-party rule in Lesotho to a perceived association between multiparty competition and political violence. In Tanzania, this same condition is explained by popular loyalty to a charismatic political leader pushing the one-party agenda. I believe my findings in the previous two chapters are in agreement with this framework and show that getting up close to each country in this way might be the best available method to discover the roots of regime support.
Central and Eastern Europe
While it is hard to generalize for Central and Eastern Europe given the small number of cases analyzed, there are some distinct trends in the region. Both economic and political variables seemed somewhat relevant to explaining authoritarian support in this region. The two countries with higher popular authoritarian support were characterized by struggling economies
(lower GDP growth and higher inflation). Level of education seems to be an important factor as well. Objective and subjective measures confirmed that poverty may be driving some in this region toward authoritarian alternatives.
112 Turning to politics, support for strongman rule was closely predicted by perceptions that rights are not respected and that elections are corrupt. This offered some of the strongest evidence that evaluations of the functioning of democracy can cause support for non-democratic alternatives in some cases. The effect of culture was unclear at the regional level but further analysis in Chapter 5 revealed that many in this region (especially in Romania and Ukraine) see both strongman rule and democracy as acceptable. Emancipative values revealed that Ukrainian culture might be more similar to Russia than Europe.
Overarching Themes
One major finding of this thesis is that authoritarian support may be best observed as distinct from democratic support. One reason for this, demonstrated in Chapter 3, is that a drop in support for democracy can simply mean increasing indifference. Poor political conditions in particular seem to drive indifference to a greater degree than approval of non-democratic alternatives. Support for authoritarianism is a comparatively extreme option and seems to have somewhat distinct sources. Additionally, the case of Ukraine in Chapter 5 demonstrated that sometimes support for democracy and support for authoritarianism can coexist without being seen as contradictory. This means that the nearly universal support for democracy in general across the world is likely deceiving.
Another key finding relates to the level of analysis most appropriate for studying support for authoritarian regime alternatives. This thesis found that different regions, and even different countries, had distinct sources of support for military rule, strong man rule and autocratic rule in general. The cases observed in Chapter 6 serve as an example of two countries that could not be explained by the traditional framework looking at economics, politics, culture or even history. In this case, the high levels of authoritarian support had more to do with the unique environment of
113 violence, insecurity and corruption that pervades Guatemala. In Central and Eastern Europe, the strong effect of political evaluations on support for strongman rule was completely absent in
Ukraine, where it seems that a hidden cultural bent was responsible for the widespread support for this authoritarian option. If these two cases were best observed at the country-level, there is good reason to think that other cases will be as well. It makes sense to investigate the issues that are close to the minds of citizens in each individual country rather than focusing too much on generalization.
114 Works Cited
Azpuru, Dinorah. “The Consulta Popular: A Vote Divided by Geography.” The Popular Referendum And The Future of the Peace Process in Guatemala . Wilson Center Working Paper 214 (1999). Booth, John A., Christine J. Wade and Thomas W. Walker. Understanding Central America: Global Forces, Rebellion and Change. Colorado: Westview Press, 2015. Booth, John A. and Patricia Bayer Richard, “Revolution’s Legacy: Residual Effects on Nicaraguan Participation and Attitudes in a Comparative Context,” Latin American Politics and Society 48(2) (2006): 117-140. Bratton, Michael and Robert Mattes. “Support for Democracy in Africa: Intrinsic or Instrumental?” British Journal of Political Science 31(3) (2001): 447-474. Bratton, Michael, Robert Mattes, and E. Gyimah-Boadi. Public Opinion, Democracy and Market Reform in Africa . New York: Cambridge University Press, 2013. “Brazil: five decades on, a key step towards truth and justice for dictatorship’s crimes.” Amnesty International , December, 10, 2014. Accessed April 15, 2015. https://www.amnesty.org/en/articles/news/2014/12/brazil-five-decades-key-step-towards- truth-and-justice-dictatorship-s-crimes/. Carreras, Miguel. “The Impact of Criminal Violence on Regime Legitimacy in Latin America.” Latin American Research Review 48(3) (2013). Chang, Yu-tzung, Yun-han Chu, and Chong-Min Park. “Authoritarian Nostalgia in Asia.” Journal of Democracy 18(3) (2007): 66-80. Chu, Yun-han, Michael Bratton, Marta Lagos, Sandeep Shastri, and Mark Tessler. “Public Opinion and Democratic Legitimacy.” Journal of Democracy 19(2) (2008). CIA. “The World Factbook.” Central Intelligence Agency , 2014. Accessed April 17, 2015. https://www.cia.gov/library/publications/the-world-factbook/geos/nu.html. Dalton, Russell. Citizen Politics: Public Opinion and Political Parties in Advanced Western Democracies. Chatham: Chatham House, 1996. Diamond, Larry. Developing Democracy: Toward Consolidation . Baltimore: The John Hopkins University Press, 1999. D’Anieri, Paul. Understanding Ukrainian Politics: Power, Politics, and Institutional Design. New York: M.E. Sharpe, Inc., 2007. Evans, Geoffrey and Stephen Whitefield. “The Politics and Economics of Democratic Commitment: Support for Democracy in Transition Societies.” British Journal of Political Science (1995): 485-514.
115 Fish, Steven M. Are Muslims Distinctive? A Look at the Evidence. New York: Oxford University Press, 2011. Freedom House. “Freedom in the World: Guatemala.” Freedom House , 2011. Accessed April 17, 2015. https://freedomhouse.org/report/freedom-world/2011/guatemala#.VTFjkpTF8vE/. Freedom House. “Freedom in the World: Nicaragua.” Freedom House , 2011. Accessed April 17, 2015. https://freedomhouse.org/report/freedom-world/2011/nicaragua#.VQpRLWTF8vE. Freedom House. “Freedom in the World: Ukraine.” Freedom House, 2011. Accessed April 17, 2015. https://freedomhouse.org/report/freedom-world/2011/ukraine#.VSckipTF8vE. Grindle, Merilee S. Despite The Odds: The Contentious Politics of Education Reform. New Jersey: Princeton University Press, 2004. Huntington, Samuel P. The Third Wave: Democratization in the Late Twentieth Century. Norman: University of Oklahoma Press, 1991. Isaacs, Anita. “Countries at the Crossroads: Guatemala.” Freedom House, 2012. Accessed April 17, 2015. https://freedomhouse.org/report/countries- crossroads/2012/guatemala#.VRCzQ2TF8vE. Karatnycky, Adrian. “Ukraine’s Orange Revolution.” Foreign Affairs 84(2) (2004): 35-52. Kitschelt, Herbert. “The Formation of Party Systems in East Central Europe.” Politics and Society (1992): 7-50. Lagos Cruz-Coke, Marta. “A Road with No Return?” Journal of Democracy 14 (2) (2003): 163- 173. Lagos Cruz-Coke, Marta. “Latin America’s Smiling Mask.” Journal of Democracy 8(3) (1997): 125-138. Latinobarómetro Corporation, 2011, "Latinobarometro / Latinobarometer Data Files", http://hdl.handle.net/1902.29/10528 Latinobarometro Corporation [Distributor] V1 [Version]. Lipset, Seymour Martin. “Some Social Requisites of Democracy: Economic Development and Political Legitimacy.” The American Political Science Review (1959): 69-105. Park, Chong-Min and Doh Chull Shin. “Do Asian Values Deter Popular Support For Democracy? The Case of South Korea.” Asian Barometer Working Paper Series 26 (2004). Pérez, Orlando J. “Democratic Legitimacy and Public Insecurity: Crime and Democracy in El Salvador and Guatemala.” Political Science Quarterly 118(4) (2004): 627-645. Person, Robert. “Popular Support for Authoritarianism: Exploring Mass Beliefs about Democracy and Dictatorship in Contemporary Ukraine.” (paper presented at the Midwest Political Science Association Annual Conference, Chicago, Illinois, April 2-5, 2009).
116 Pico, Juan Hernandez. “Why Was The Referendum Defeated?” Envio 216 (1999). Przeworski, Adam. Democracy and the Market: Political and Economic Reforms in Eastern Europe and Latin America. New York: Cambridge University Press, 1991. Reisinger, William M., Arthur H. Miller, Vicki L. Hesli and Kristen Hill Maher. “Political Values in Russia, Ukraine and Lithuania: Sources and Implications for Democracy.” British Journal of Political Science 24(2) (1994): 183-223. Rose, Richard, William Mishler, and Christian Haerpfer. Democracy and its Alternatives: Understanding Post-Communist Societies. Baltimore: John Hopkins University Press, 1998. Seligson, Mitchell A. “The Impact of Corruption on Regime Legitimacy: A Comparative Study of Four Latin American Countries.” The Journal of Politics 64(2) (2002): 408-433. Sherman, Mark. “Guatemalan Anti-Drug Official Indicted.” The Washington Post. November 16, 2005. Accessed March 23, 2015. http://www.washingtonpost.com/wp- dyn/content/article/2005/11/16/AR2005111601349.html. Transparency International. “Corruption Perceptions Index: Ukraine.” Transparency International , 2014. Accessed April 11, 2015. http://www.transparency.org/country#UKR. Vanden, Harry E. and Gary Prevost. Politics of Latin America: The Power Game . New York: Oxford University Press, 2006. Weeks, Gregory. Understanding Latin American Politics. New York: Pearson, 2015. Welzel, Christian and Ronald F. Inglehart, “Political Culture, Mass Beliefs, and Value Change.” Democratization, edited by Christian W. Haerpfer, Patrick Bernhagen, Ronald F. Inglehart, and Christian Welzel, 126-144. New York: Oxford University Press, 2009. Weyland, Kurt. “The Politics of Corruption in Latin America.” Journal of Democracy 9(2) (1998): 108-121. Wiarda, Howard, ed. Politics and Social Change in Latin America: The Distinct Tradition. Amherst: University of Massachusetts Press, 1974. Wolchik, Sharon L. and Jane Leftwich Curry, ed. Central and East European Politics: From Communism to Democracy . Lanham: Rowman & Littlefield, 2015. The World Bank. “World Development Indicators.” The World Bank , 2012. Accessed on April 17, 2015. http://data.worldbank.org/indicator/SI.POV.NAHC/countries?page=1. WORLD VALUES SURVEY Wave 6 2010-2014 OFFICIAL AGGREGATE v.20141107. World Values Survey Association (www.worldvaluessurvey.org). Aggregate File Producer: Asep/JDS, Madrid SPAIN.
117